当前位置:   article > 正文

ICCV2019论文题目中文列表_deep restoration of vintage photographs from scann

deep restoration of vintage photographs from scanned halftone prints
英文题目中文题目 
FaceForensics++: Learning to Detect Manipulated Facial ImagesFaceForensics++:学习检测操纵的面部图像 
DeepVCP: An End-to-End Deep Neural Network for Point Cloud RegistrationDeepVCP:用于点云配准端到端深度神经网络 
Shape Reconstruction Using Differentiable Projections and Deep Priors基于可微投影深度先验形状重建 
Fine-Grained Segmentation Networks: Self-Supervised Segmentation for Improved Long-Term Visual Localization细粒度分割网络:基于自监督分割长期视觉定位性能提升 
SANet: Scene Agnostic Network for Camera LocalizationSANet:基于场景不可知网络摄像机定位 
Total Denoising: Unsupervised Learning of 3D Point Cloud Cleaning全消噪:三维点云清理无监督学习 
Hierarchical Self-Attention Network for Action Localization in Videos视频动作定位分层自关注网络 
Goal-Driven Sequential Data Abstraction目标驱动顺序数据抽象 
Jointly Aligning Millions of Images With Deep Penalised Reconstruction Congealing基于深度惩罚重建凝结数百万张图片联合对齐 
Drop to Adapt: Learning Discriminative Features for Unsupervised Domain Adaptation放弃适应:基于判别特征学习非监督域适应 
NLNL: Negative Learning for Noisy LabelsNLNL:噪声标签负学习 
Adversarial Robustness vs. Model Compression, or Both?对抗性稳健Vs.模型压缩,或两者兼而有之? 
On the Design of Black-Box Adversarial Examples by Leveraging Gradient-Free Optimization and Operator Splitting Method利用无梯度优化算子分裂方法设计黑盒对抗实例 
DewarpNet: Single-Image Document Unwarping With Stacked 3D and 2D Regression NetworksDewarpNet:使用叠加三维和二维回归网络单图像文档去弯曲 
Learning Robust Facial Landmark Detection via Hierarchical Structured Ensemble基于层次结构集成鲁棒人脸Landmark检测 
Remote Heart Rate Measurement From Highly Compressed Facial Videos: An End-to-End Deep Learning Solution With Video Enhancement高度压缩的面部视频中进行远程心率测量:一种具有视频增强功能端到端深度学习解决方案 
Face-to-Parameter Translation for Game Character Auto-Creation面向参数转换游戏角色自动生成 
Visual Deprojection: Probabilistic Recovery of Collapsed Dimensions可视化反投影坍塌维度的概率恢复 
StructureFlow: Image Inpainting via Structure-Aware Appearance Flow结构流:通过结构感知外观流进行图像修复 
Learning Fixed Points in Generative Adversarial Networks: From Image-to-Image Translation to Disease Detection and LocalizationGAN中的不动点学习:从图像-图像转换疾病检测与定位 
Generative Adversarial Training for Weakly Supervised Cloud Matting基于生成性对抗训练弱监督云Matting(抠图?) 
PAMTRI: Pose-Aware Multi-Task Learning for Vehicle Re-Identification Using Highly Randomized Synthetic DataPAMTRI:基于高度随机综合数据姿态感知多任务学习实现车辆再识别 
Generative Adversarial Networks for Extreme Learned Image Compression用于极端学习图像压缩GAN 
Instance-Guided Context Rendering for Cross-Domain Person Re-Identification基于实例引导上下文呈现跨域人再识别 
What Else Can Fool Deep Learning? Addressing Color Constancy Errors on Deep Neural Network Performance还有什么可以愚弄深度学习?深度神经网络性能色彩恒常性误差处理 
Beyond Cartesian Representations for Local Descriptors超越笛卡尔表示局部描述符 
Distilling Knowledge From a Deep Pose Regressor Network深度姿态回归网络提取知识 
Instance-Level Future Motion Estimation in a Single Image Based on Ordinal Regression基于序贯回归单帧图像实例级未来运动估计 
Vision-Infused Deep Audio Inpainting视觉注入深度音频修复 
HAWQ: Hessian AWare Quantization of Neural Networks With Mixed-PrecisionHAWQ:利用混合精度实现神经网络的Hessian感知量化 
Evaluating Robustness of Deep Image Super-Resolution Against Adversarial Attacks深度图像超分辨率抗对抗攻击鲁棒性评估 
Overcoming Catastrophic Forgetting With Unlabeled Data in the Wild利用野外无标签数据克服灾难性遗忘 
Symmetric Cross Entropy for Robust Learning With Noisy Labels带噪声标签鲁棒学习对称交叉熵 
Few-Shot Learning With Embedded Class Models and Shot-Free Meta Training基于嵌入式类模型无镜头元训练少镜头学习 
Dual Directed Capsule Network for Very Low Resolution Image Recognition用于超低分辨率图像识别双向胶囊网络 
Recognizing Part Attributes With Insufficient Data利用不足数据识别部分属性 
USIP: Unsupervised Stable Interest Point Detection From 3D Point CloudsUSIP:三维点云无监督稳定兴趣点检测 
Mixed High-Order Attention Network for Person Re-Identification混合高阶注意网络用于人再识别 
Budget-Aware Adapters for Multi-Domain Learning用于多域学习预算感知适配器 
Compact Trilinear Interaction for Visual Question Answering视觉问答紧凑三线交互 
Towards Latent Attribute Discovery From Triplet Similarities基于三元相似性潜在属性发现 
GeoStyle: Discovering Fashion Trends and EventsGeoStyle:发现时尚趋势事件 
Towards Adversarially Robust Object Detection对抗性鲁棒目标检测 
Automatic and Robust Skull Registration Based on Discrete Uniformization基于离散均匀化自动鲁棒颅骨配准 
Few-Shot Image Recognition With Knowledge Transfer基于知识迁移少镜头图像识别 
Fine-Grained Action Retrieval Through Multiple Parts-of-Speech Embeddings基于多重部分语音嵌入细粒度动作检索 
Vehicle Re-Identification in Aerial Imagery: Dataset and Approach航空影像中的车辆再识别数据集方法 
Bridging the Domain Gap for Ground-to-Aerial Image Matching地-空图像匹配中的域间隙桥接 
A Robust Learning Approach to Domain Adaptive Object Detection一种鲁棒学习域自适应目标检测方法 
Graph-Based Object Classification for Neuromorphic Vision Sensing基于对象分类实现神经形态视觉感知 
Gaussian YOLOv3: An Accurate and Fast Object Detector Using Localization Uncertainty for Autonomous Driving高斯YOLOv3:自主驾驶中一种基于定位不确定性快速精确目标检测方法 
Sharpen Focus: Learning With Attention Separability and Consistency集中注意力:学习中注意力可分离性一致性 
Learning Semantic-Specific Graph Representation for Multi-Label Image Recognition多标签图像识别中的特定语义图表示学习 
DeceptionNet: Network-Driven Domain RandomizationDeceptionNet:网络驱动域随机化 
Pose-Guided Feature Alignment for Occluded Person Re-Identification基于姿态引导特征对齐实现遮挡人再识别 
Robust Person Re-Identification by Modelling Feature Uncertainty基于特征不确定性建模鲁棒人再识别 
Co-Segmentation Inspired Attention Networks for Video-Based Person Re-Identification基于共分割启发注意网络用于基于视频的人再识别 
A Delay Metric for Video Object Detection: What Average Precision Fails to Tell视频目标检测一种延迟度量:平均精度不能判断什么 
IL2M: Class Incremental Learning With Dual MemoryIL2M:双记忆课堂增量学习 
Asymmetric Non-Local Neural Networks for Semantic Segmentation非对称非局部神经网络用于语义分割语义分割网中嵌入NonLocal-Block,并将其改进为非对称NonLocal-Block,并进一步添加金字塔池化和多级融合技术(见框图)
CCNet: Criss-Cross Attention for Semantic SegmentationCCNet:基于交叉注意语义分割利用十字(criss-cross)方式,高效地获取全局上下文信息
Convex Shape Prior for Multi-Object Segmentation Using a Single Level Set Function基于单水平集函数凸形状先验多目标分割 
Feature Weighting and Boosting for Few-Shot Segmentation基于特征加权boosting少镜头分割 
Surface Networks via General Covers通过一般覆盖地面网络 
SSAP: Single-Shot Instance Segmentation With Affinity PyramidSSAP:基于相似金字塔单镜头实例分割先进行(多尺度)语义分割(S),同时获得(多尺度)像素对关系(A),最后将不同尺度的A和S利用图割的方式,融合在一起,得到实例分割
Learning Propagation for Arbitrarily-Structured Data面向任意结构数据学习传播 
MultiSeg: Semantically Meaningful, Scale-Diverse Segmentations From Minimal User InputMultiSeg:从最小的用户输入实现语义上有意义,尺度分散的分割 
Robust Motion Segmentation From Pairwise Matches基于成对匹配鲁棒运动分割 
InstaBoost: Boosting Instance Segmentation via Probability Map Guided Copy-PastingInstaBoost:通过概率地图引导复制粘贴实现增强实例分割利用Copy-Paste的方法,实现训练样本集的增强
Racial Faces in the Wild: Reducing Racial Bias by Information Maximization Adaptation Network荒野中的种族面孔:信息最大化自适应网络减少种族偏移 
Uncertainty Modeling of Contextual-Connections Between Tracklets for Unconstrained Video-Based Face RecognitionTracklet间上下文关系不确定性建模实现无约束视频人脸识别 
Spatio-Temporal Fusion Based Convolutional Sequence Learning for Lip Reading基于时空融合卷积序列学习实现唇读 
Occlusion-Aware Networks for 3D Human Pose Estimation in Video视频中三维人体姿态估计遮挡感知网络 
Context-Aware Feature and Label Fusion for Facial Action Unit Intensity Estimation With Partially Labeled Data利用部分标记数据实现基于上下文感知特征标签融合人脸动作单元强度估计 
Distill Knowledge From NRSfM for Weakly Supervised 3D Pose Learning基于NRSfM知识蒸馏实现弱监督三维姿态学习 
MONET: Multiview Semi-Supervised Keypoint Detection via Epipolar Divergence基于极线散度多视图半监督关键点检测 
Talking With Hands 16.2M: A Large-Scale Dataset of Synchronized Body-Finger Motion and Audio for Conversational Motion Analysis and Synthesis手语16.2m:用于会话运动分析和合成大规模体-指同步运动和音频数据集 
Occlusion Robust Face Recognition Based on Mask Learning With Pairwise Differential Siamese Network基于成对差分孪生网络模板学习实现遮挡鲁棒人脸识别 
Teacher Supervises Students How to Learn From Partially Labeled Images for Facial Landmark Detection教师指导学生如何从部分标记的图像学习面部Landmark检测 
A2J: Anchor-to-Joint Regression Network for 3D Articulated Pose Estimation From a Single Depth ImageA2J:锚定联合回归网络用于单深度图像三维关节姿态估计 
TexturePose: Supervising Human Mesh Estimation With Texture Consistency基于纹理一致性人体网格估计监控 
FreiHAND: A Dataset for Markerless Capture of Hand Pose and Shape From Single RGB ImagesFreiHAND:一个从单个RGB图像无标记捕捉手部姿势形状数据集 
Markerless Outdoor Human Motion Capture Using Multiple Autonomous Micro Aerial Vehicles多自主微型飞行器无标记室外人体运动捕捉 
Toyota Smarthome: Real-World Activities of Daily Living丰田智能家居:现实生活中的日常生活活动 
Relation Parsing Neural Network for Human-Object Interaction Detection关系解析神经网络人机交互检测中的应用 
DistInit: Learning Video Representations Without a Single Labeled VideoDistInit:学习没有单个标记视频视频表示 
Zero-Shot Anticipation for Instructional Activities教学活动的零镜头预期 
Making the Invisible Visible: Action Recognition Through Walls and Occlusions使隐形可见:通过墙和遮挡动作识别 
Recursive Visual Sound Separation Using Minus-Plus NetMinus-Plus Net进行递归可视声音分离 
Unsupervised Video Interpolation Using Cycle Consistency基于循环一致性无监督视频插值 
Deformable Surface Tracking by Graph Matching基于图匹配变形曲面跟踪 
Deep Meta Learning for Real-Time Target-Aware Visual Tracking基于深度元学习实时目标感知视觉跟踪 
Looking to Relations for Future Trajectory Forecast展望未来轨迹预测关系 
Anchor Diffusion for Unsupervised Video Object Segmentation无监督视频对象分割锚扩散算法 
Tracking Without Bells and Whistles无铃无哨的追踪 
Perspective-Guided Convolution Networks for Crowd Counting面向人群计数透视导引卷积网络 
End-to-End Wireframe Parsing端到端线框分析 
Incremental Class Discovery for Semantic Segmentation With RGBD Sensing基于RGBD感知增量类发现实现语义分割 
SSF-DAN: Separated Semantic Feature Based Domain Adaptation Network for Semantic SegmentationSSF-DAN:基于分离语义特征域自适应实现语义分割(待标签的)训练样本与真实域无标签训练样本在不同域,因此采用域自适应的方法,来实现弱监督的语义分割。本文采用GAN的方法,如图2
SpaceNet MVOI: A Multi-View Overhead Imagery DatasetSpaceNet-MVOI:一个多视图俯视图像数据集 
Multi-Level Bottom-Top and Top-Bottom Feature Fusion for Crowd Counting用于人群计数多层次自下而上和自上而下特征融合 
Learning Lightweight Lane Detection CNNs by Self Attention Distillation自关注蒸馏学习轻量级CNNs用于车道检测 
SplitNet: Sim2Sim and Task2Task Transfer for Embodied Visual NavigationSplitNet:Sim2SimTask2Task传输以实现可视化导航 
Cascaded Parallel Filtering for Memory-Efficient Image-Based Localization基于级联并行滤波记忆高效图像定位 
Pixel2Mesh++: Multi-View 3D Mesh Generation via DeformationPixel2Mesh++:通过变形生成多视图三维网格 
A Differential Volumetric Approach to Multi-View Photometric Stereo基于差分体积法多视光度立体成像 
Revisiting Radial Distortion Absolute Pose重新审视径向畸变绝对姿态 
Estimating the Fundamental Matrix Without Point Correspondences With Application to Transmission Imaging无点对应的基本矩阵估计及其在透射成像中的应用 
QUARCH: A New Quasi-Affine Reconstruction Stratum From Vague Relative Camera Orientation KnowledgeQUARCH:一种基于模糊相对摄像机方位知识准仿射重建层 
Homography From Two Orientation- and Scale-Covariant Features基于两个方向尺度协方差特征单应性 
Hiding Video in Audio via Reversible Generative Models基于可逆生成模型隐藏视频到音频 
GSLAM: A General SLAM Framework and BenchmarkGSLAM:一个通用的SLAM框架基准 
Elaborate Monocular Point and Line SLAM With Robust Initialization具有鲁棒初始化精细单目点-线SLAM 
Adaptive Density Map Generation for Crowd Counting用于人群计数自适应密度图生成 
Attention-Aware Polarity Sensitive Embedding for Affective Image Retrieval注意力感知极性敏感嵌入情感图像检索中的应用 
Zero-Shot Emotion Recognition via Affective Structural Embedding基于情感结构嵌入零镜头情感识别 
FW-GAN: Flow-Navigated Warping GAN for Video Virtual Try-OnFW-GAN:用于视频虚拟试穿流导航翘曲GAN 
Interactive Sketch & Fill: Multiclass Sketch-to-Image Translation交互式草图与填充多类别草图-图像转换 
Attention-Based Autism Spectrum Disorder Screening With Privileged Modality基于注意力自闭症谱系障碍筛查 
Image Aesthetic Assessment Based on Pairwise Comparison A Unified Approach to Score Regression, Binary Classification, and Personalization基于成对比较的图像美学评价评分回归二元分类个性化统一方法 
Delving Into Robust Object Detection From Unmanned Aerial Vehicles: A Deep Nuisance Disentanglement Approach无人机鲁棒目标检测的深入研究 
Bit-Flip Attack: Crushing Neural Network With Progressive Bit Search比特翻转攻击:基于渐进式比特搜索粉碎神经网络 
Pushing the Frontiers of Unconstrained Crowd Counting: New Dataset and Benchmark Method推动无约束人群计数的前沿:新数据集基准方法 
Employing Deep Part-Object Relationships for Salient Object Detection利用深度局部-目标关系进行显著目标检测 
Self-Supervised Deep Depth Denoising自监督深度深度信息去噪 
Cost-Aware Fine-Grained Recognition for IoTs Based on Sequential Fixations成本感知细粒度识别实现顺序固定的IoT(物联网?) 
Layout-Induced Video Representation for Recognizing Agent-in-Place Actions基于布局诱导的视频表示方法识别Agent原地动作 
Anomaly Detection in Video Sequence With Appearance-Motion Correspondence基于外观运动对应视频序列异常检测 
Exploring Randomly Wired Neural Networks for Image Recognition随机有线神经网络图像识别中的应用 
Progressive Differentiable Architecture Search: Bridging the Depth Gap Between Search and Evaluation渐进可微架构搜索缩小搜索评估之间的深度差距 
Multinomial Distribution Learning for Effective Neural Architecture Search基于多项式分布学习的有效的神经结构搜索 
Searching for MobileNetV3正在搜索MobileNetV3 
Data-Free Quantization Through Weight Equalization and Bias Correction通过权值均衡偏差校正实现无数据量化 
A Camera That CNNs: Towards Embedded Neural Networks on Pixel Processor ArraysCNNs摄像机:面向像素处理器阵列上的嵌入式神经网络 
Knowledge Distillation via Route Constrained Optimization基于路径约束优化知识蒸馏 
Distillation-Based Training for Multi-Exit Architectures基于蒸馏训练实现多出口结构 
Similarity-Preserving Knowledge Distillation相似性保持知识蒸馏 
Many Task Learning With Task Routing基于任务路由多任务学习 
Stochastic Filter Groups for Multi-Task CNNs: Learning Specialist and Generalist Convolution Kernels基于随机滤波器组多任务CNN学习专家广义卷积核 
Transferability and Hardness of Supervised Classification Tasks监督分类任务可转移性难易性 
Moment Matching for Multi-Source Domain Adaptation基于矩匹配多源域自适应 
Unsupervised Domain Adaptation via Regularized Conditional Alignment基于正则条件对齐无监督域自适应 
Larger Norm More Transferable: An Adaptive Feature Norm Approach for Unsupervised Domain Adaptation更大范数更多可转移:一种无监督域自适应自适应特征范数方法 
UM-Adapt: Unsupervised Multi-Task Adaptation Using Adversarial Cross-Task DistillationUM-Adapt:使用对抗性跨任务蒸馏无监督多任务自适应 
Episodic Training for Domain Generalization基于幕式训练域泛化 
Domain Adaptation for Structured Output via Discriminative Patch Representations基于判别区分块表示结构化输出域自适应 
Semi-Supervised Learning by Augmented Distribution Alignment基于增广分布对齐半监督学习 
S4L: Self-Supervised Semi-Supervised LearningS4L:自监督半监督学习 
Privacy Preserving Image Queries for Camera Localization隐私保护图像查询实现摄像机定位 
Calibration Wizard: A Guidance System for Camera Calibration Based on Modelling Geometric and Corner Uncertainty标定向导:一种基于几何不确定性建模摄像机标定制导系统 
Gated2Depth: Real-Time Dense Lidar From Gated ImagesGated2Depth:来自门控图像实时密集激光雷达 
X-Section: Cross-Section Prediction for Enhanced RGB-D Fusionx截面:增强RGBD融合截面预测 
Learning an Event Sequence Embedding for Dense Event-Based Deep Stereo事件序列嵌入学习实现基于稠密事件的深度立体图 
Point-Based Multi-View Stereo Network基于多视图立体网络 
Discrete Laplace Operator Estimation for Dynamic 3D Reconstruction动态三维重建离散Laplace算子估计 
Deep Non-Rigid Structure From Motion深度非刚性Structure From Motion 
Equivariant Multi-View Networks等变多视网络 
Interpolated Convolutional Networks for 3D Point Cloud Understanding插值卷积网络三维点云理解中的应用 
Revisiting Point Cloud Classification: A New Benchmark Dataset and Classification Model on Real-World Data重新审视点云分类:一种基于真实数据的新的基准数据集分类模型 
PointCloud Saliency Maps点云显著图 
ShellNet: Efficient Point Cloud Convolutional Neural Networks Using Concentric Shells Statistics基于同心壳统计的高效点云卷积神经网络 
Unsupervised Deep Learning for Structured Shape Matching基于无监督深度学习结构形状匹配 
Linearly Converging Quasi Branch and Bound Algorithms for Global Rigid Registration基于线性收敛准分枝定界算法的全局刚性配准 
Consensus Maximization Tree Search Revisited协商一致最大化树搜索 
Quasi-Globally Optimal and Efficient Vanishing Point Estimation in Manhattan World曼哈顿世界的准全局最优高效消失点估计 
An Efficient Solution to the Homography-Based Relative Pose Problem With a Common Reference Direction具有共同参考方向单应相对位姿问题有效解 
A Quaternion-Based Certifiably Optimal Solution to the Wahba Problem With Outliers基于四元数孤立点Wahba问题可证明最优解 
PLMP - Point-Line Minimal Problems in Complete Multi-View Visibility完全多视图可见性中的点-线最小问题 
Variational Few-Shot Learning变分少镜头学习 
Generative Adversarial Minority Oversampling生成性对抗少数过采样 
Memorizing Normality to Detect Anomaly: Memory-Augmented Deep Autoencoder for Unsupervised Anomaly Detection记忆正态性异常检测:用于无监督异常检测记忆增强深度自动编码器 
Topological Map Extraction From Overhead Images从头顶图像提取拓扑图 
Exploiting Temporal Consistency for Real-Time Video Depth Estimation利用时间一致性进行实时视频深度估计 
The Sound of Motions运动的声音 
SC-FEGAN: Face Editing Generative Adversarial Network With User's Sketch and ColorSC-FEGAN:基于用户素描色彩人脸编辑生成对抗网络 
Exploring Overall Contextual Information for Image Captioning in Human-Like Cognitive Style探索类人认知方式图像字幕整体语境信息 
Order-Aware Generative Modeling Using the 3D-Craft Dataset基于三维工艺数据集次序感知生成建模 
Crowd Counting With Deep Structured Scale Integration Network基于深度结构规模集成网络人群计数 
Bidirectional One-Shot Unsupervised Domain Mapping双向单镜头无监督域映射 
Evolving Space-Time Neural Architectures for Videos进化的视频时空神经结构 
Universally Slimmable Networks and Improved Training Techniques通用可瘦身网络改进的训练技术 
AutoDispNet: Improving Disparity Estimation With AutoMLAutoDispNet:用AutoML改进视差估计网络结构搜索和最优超参数搜索的方法
Deep Meta Functionals for Shape Representation基于深度元函数形状表示 
Differentiable Kernel Evolution可微的核演化 
Batch Weight for Domain Adaptation With Mass Shift利用质量漂移实现域自适应批处权重 
SRM: A Style-Based Recalibration Module for Convolutional Neural NetworksSRM:卷积神经网络中一种基于样式再校准模块 
Switchable Whitening for Deep Representation Learning基于可切换白化深度表示学习 
Adaptative Inference Cost With Convolutional Neural Mixture Models基于卷积神经混合模型自适应推理代价 
On Network Design Spaces for Visual Recognition基于网络设计空间视觉识别 
Improved Techniques for Training Adaptive Deep Networks自适应深度网络训练改进技术 
Resource Constrained Neural Network Architecture Search: Will a Submodularity Assumption Help?资源受限的神经网络架构搜索子模块假设有帮助吗? 
ACNet: Strengthening the Kernel Skeletons for Powerful CNN via Asymmetric Convolution BlocksACNet:通过非对称卷积块增强CNN的核骨架 
A Comprehensive Overhaul of Feature Distillation特征蒸馏全面检修 
Transferable Semi-Supervised 3D Object Detection From RGB-D DataRGBD数据可转移半监督三维目标检测 
DPOD: 6D Pose Object Detector and RefinerDPOD:6D位姿目标检测器细化器 
STD: Sparse-to-Dense 3D Object Detector for Point CloudSTD:点云稀疏-稠密三维目标检测器 
DUP-Net: Denoiser and Upsampler Network for 3D Adversarial Point Clouds DefenseDUP-Net:用于3D对抗点云防御去噪和上采样网络 
Learning Rich Features at High-Speed for Single-Shot Object Detection高速学习丰富特征实现单镜头目标检测 
Detecting Unseen Visual Relations Using Analogies类比法检测看不见的视觉关系 
Disentangling Monocular 3D Object Detection分离式单目三维目标检测 
STM: SpatioTemporal and Motion Encoding for Action RecognitionSTM:用于动作识别时空和运动编码 
Dynamic Context Correspondence Network for Semantic Alignment语义对齐动态上下文对应网络 
Fooling Network Interpretation in Image Classification图像分类中的愚弄网络解释 
Unconstrained Foreground Object Search无约束前景对象搜索 
Embodied Amodal Recognition: Learning to Move to Perceive Objects体现性情感识别学习移动感知物体 
SpatialSense: An Adversarially Crowdsourced Benchmark for Spatial Relation Recognition空间感知:一种用于空间关系识别逆向众包基准 
TensorMask: A Foundation for Dense Object SegmentationTensorMask:密集目标分割基础 
Integral Object Mining via Online Attention Accumulation基于在线注意力积累整体对象挖掘 
Accelerated Gravitational Point Set Alignment With Altered Physical Laws改变的物理定律加速引力点集对准 
Domain Adaptation for Semantic Segmentation With Maximum Squares Loss基于最大平方损失域自适应实现语义分割基于域自适应的语义分割,提出两点改进:1. 提出新的损失函数;2. 提出类别重加权,以解决类别不平衡的问题
Domain Randomization and Pyramid Consistency: Simulation-to-Real Generalization Without Accessing Target Domain Data域随机化金字塔一致性不访问目标域数据真实综合仿真 
Semi-Supervised Skin Detection by Network With Mutual Guidance基于互导网络半监督皮肤检测 
ACE: Adapting to Changing Environments for Semantic SegmentationACE:适应不断变化的环境实现语义分割基于域自适应的语义分割
Efficient Segmentation: Learning Downsampling Near Semantic Boundaries有效分割:在语义边界附近学习下采样 
Recurrent U-Net for Resource-Constrained Segmentation基于递归U-Net资源受限分割 
Detecting the Unexpected via Image Resynthesis通过图像再合成检测意外 
Self-Supervised Monocular Depth Hints自监督单目深度提示 
3D Scene Reconstruction With Multi-Layer Depth and Epipolar Transformers基于多层深度极线变换三维场景重建 
How Do Neural Networks See Depth in Single Images?神经网络如何在单个图像看到深度 
On Boosting Single-Frame 3D Human Pose Estimation via Monocular Videos单目视频增强单帧三维人体姿态估计 
Canonical Surface Mapping via Geometric Cycle Consistency基于几何循环一致性正则曲面映射 
3D-RelNet: Joint Object and Relational Network for 3D Prediction3d RelNet:三维预测联合对象和关系网络 
GP2C: Geometric Projection Parameter Consensus for Joint 3D Pose and Focal Length Estimation in the WildGP2C:基于几何投影参数一致性的野外联合三维姿态焦距估计 
Moulding Humans: Non-Parametric 3D Human Shape Estimation From Single Images塑造人:基于单个图像非参数三维人体形状估计 
3DPeople: Modeling the Geometry of Dressed Humans3DPeople:为穿着衣服的人的几何体建模 
Learning to Reconstruct 3D Human Pose and Shape via Model-Fitting in the Loop基于模型拟合三维人体姿态形状重建 
Optimizing Network Structure for 3D Human Pose Estimation三维人体姿态估计网络结构优化 
Exploiting Spatial-Temporal Relationships for 3D Pose Estimation via Graph Convolutional Networks基于时-空关系图形卷积网络实现三维姿态估计 
Resolving 3D Human Pose Ambiguities With 3D Scene Constraints利用三维场景约束解决三维人体姿态模糊问题 
Tex2Shape: Detailed Full Human Body Geometry From a Single ImageTex2Shape:从一幅图像获得详细的全身几何图形 
PIFu: Pixel-Aligned Implicit Function for High-Resolution Clothed Human DigitizationPIFu:基于像素对齐隐函数高分辨率服装数字化 
DF2Net: A Dense-Fine-Finer Network for Detailed 3D Face ReconstructionDF2Net:一种密集-精细-更精细网络实现详细三维人脸重建 
Monocular 3D Human Pose Estimation by Generation and Ordinal Ranking基于生成序数排序单目三维人体姿态估计 
Aligning Latent Spaces for 3D Hand Pose Estimation基于潜在空间对齐三维手部姿态估计 
HEMlets Pose: Learning Part-Centric Heatmap Triplets for Accurate 3D Human Pose EstimationHEMLets Pose:学习以局部为中心的热图三元组精确估计三维人体姿势 
End-to-End Hand Mesh Recovery From a Monocular RGB Image单目RGB图像端到端手部网格恢复 
Robust Multi-Modality Multi-Object Tracking鲁棒多模态多目标跟踪 
The Trajectron: Probabilistic Multi-Agent Trajectory Modeling With Dynamic Spatiotemporal Graphs基于动态时空图概率多智能体轨迹建模 
'Skimming-Perusal' Tracking: A Framework for Real-Time and Robust Long-Term Tracking“略读”跟踪:一个实时健壮长期跟踪框架 
TASED-Net: Temporally-Aggregating Spatial Encoder-Decoder Network for Video Saliency Detection用于视频显著性检测时间聚集空间编解码网络 
Attacking Optical Flow攻击光流 
Pro-Cam SSfM: Projector-Camera System for Structure and Spectral Reflectance From MotionPro-Cam SSfm:用于运动中结构光谱反射投影-摄像系统 
Mop Moire Patterns Using MopNet基于MopNetMop Moire图案 
Kernel Modeling Super-Resolution on Real Low-Resolution Images真实低分辨率图像核模型超分辨率 
Learning to Jointly Generate and Separate Reflections学会共同产生分离反射 
Deep Multi-Model Fusion for Single-Image Dehazing基于深度多模型融合单图像去雾 
Deep Learning for Seeing Through Window With Raindrops透过雨滴看窗外深度学习 
Mask-ShadowGAN: Learning to Remove Shadows From Unpaired DataMask-ShadowGAN学习从未配对数据移除阴影 
Spatio-Temporal Filter Adaptive Network for Video Deblurring用于视频去模糊时空滤波自适应网络 
Learning Deep Priors for Image Dehazing图像去模糊深度先验学习 
JPEG Artifacts Reduction via Deep Convolutional Sparse Coding基于深度卷积稀疏编码jpeg伪影抑制 
Self-Guided Network for Fast Image Denoising用于快速图像去噪自引导网络 
Non-Local Intrinsic Decomposition With Near-Infrared Priors基于近红外先验非局部本征分解 
VideoMem: Constructing, Analyzing, Predicting Short-Term and Long-Term Video MemorabilityVideoMem:构建分析预测短期和长期视频记忆 
Rescan: Inductive Instance Segmentation for Indoor RGBD ScansRescan:基于归纳实例分割室内RGBD扫描 
End-to-End CAD Model Retrieval and 9DoF Alignment in 3D Scans三维扫描中的端到端CAD模型检索9自由度对准 
Making History Matter: History-Advantage Sequence Training for Visual Dialog创造历史:基于历史优势序列训练可视化对话 
Stochastic Attraction-Repulsion Embedding for Large Scale Image Localization随机吸引-排斥嵌入大规模图像定位中的应用 
Scene Graph Prediction With Limited Labels基于有限标签场景图预测 
Taking a HINT: Leveraging Explanations to Make Vision and Language Models More Grounded提示:利用解释使视觉和语言模型更加扎根 
Align2Ground: Weakly Supervised Phrase Grounding Guided by Image-Caption AlignmentAlign2Ground:图片-标注对齐引导弱监督phase groundingphrase grounding:给出一张图片和一个自然语言描述的问题,在图片中定位问题中所提到的物体。是很多问题的
Adaptive Reconstruction Network for Weakly Supervised Referring Expression Grounding基于自适应重构网络弱监督指代表达 
Hierarchy Parsing for Image Captioning基于层次分析图像标注 
HowTo100M: Learning a Text-Video Embedding by Watching Hundred Million Narrated Video ClipsHowTo100M:通过观看一亿个叙述视频片段实现文本-视频嵌入学习 
Controllable Video Captioning With POS Sequence Guidance Based on Gated Fusion Network基于门控融合网络POS序列引导实现可控视频标注 
Multi-View Stereo by Temporal Nonparametric Fusion基于时间非参数融合多视点立体视觉 
Floor-SP: Inverse CAD for Floorplans by Sequential Room-Wise Shortest PathFloor-SP:按顺序房间最短路径进行楼层平面逆向CAD 
Polarimetric Relative Pose Estimation极化相对位姿估计 
Closed-Form Optimal Two-View Triangulation Based on Angular Errors基于角度误差闭式最优二视图三角剖分 
Pix2Vox: Context-Aware 3D Reconstruction From Single and Multi-View ImagesPix2Vox:基于单视图多视图图像上下文感知三维重建 
Unsupervised Robust Disentangling of Latent Characteristics for Image Synthesis潜在特征的无监督鲁棒分离实现图像合成 
SROBB: Targeted Perceptual Loss for Single Image Super-ResolutionSROBB:单图像超分辨率目标感知损失 
An Internal Learning Approach to Video Inpainting视频修复内部学习方法 
Deep CG2Real: Synthetic-to-Real Translation via Image Disentanglement深层CG2Real:通过图像解纠缠实现从合成到真实的翻译 
Adversarial Defense via Learning to Generate Diverse Attacks通过学习产生多种攻击实现对抗性防御 
Image Generation From Small Datasets via Batch Statistics Adaptation批统计自适应实现从小数据集生成图像 
Lifelong GAN: Continual Learning for Conditional Image Generation终身GAN条件图像生成持续学习 
Bayesian Relational Memory for Semantic Visual Navigation面向语义视觉导航贝叶斯关系记忆 
Mono-SF: Multi-View Geometry Meets Single-View Depth for Monocular Scene Flow Estimation of Dynamic Traffic ScenesMono-SF:多视点几何满足单视点深度单目动态交通场景流量估计 
Prior Guided Dropout for Robust Visual Localization in Dynamic Environments基于先验引导Dropout动态环境鲁棒视觉定位 
Drive&Act: A Multi-Modal Dataset for Fine-Grained Driver Behavior Recognition in Autonomous VehiclesDrive&Act:一个用于自主车辆细粒度驾驶员行为识别多模态数据集 
Depth Completion From Sparse LiDAR Data With Depth-Normal Constraints基于深度法向约束稀疏激光雷达数据深度补全 
PRECOG: PREdiction Conditioned on Goals in Visual Multi-Agent SettingsPRECOG:视觉多Agent设置基于目标的预测 
LPD-Net: 3D Point Cloud Learning for Large-Scale Place Recognition and Environment AnalysisLPD-Net:用于大规模地点识别环境分析三维点云学习 
Local Supports Global: Deep Camera Relocalization With Sequence Enhancement局部支持全局:基于序列增强深度相机重定位 
Sequential Adversarial Learning for Self-Supervised Deep Visual Odometry基于序贯对抗学习自监督深度视觉里程计 
TextPlace: Visual Place Recognition and Topological Localization Through Reading Scene Texts文本位置:通过阅读场景文本进行视觉位置识别拓扑定位 
CamNet: Coarse-to-Fine Retrieval for Camera Re-LocalizationCamNet:从粗到细的检索实现相机重定位 
Situational Fusion of Visual Representation for Visual Navigation视觉表示情景融合实现视觉导航 
Learning Aberrance Repressed Correlation Filters for Real-Time UAV Tracking学习畸变抑制相关滤波器无人机实时跟踪中的应用 
6-DOF GraspNet: Variational Grasp Generation for Object Manipulation六自由度GraspNet:基于变分抓取生成对象操作 
DAGMapper: Learning to Map by Discovering Lane TopologyDAGMapper:通过发现车道拓扑学习地图 
3D-LaneNet: End-to-End 3D Multiple Lane Detection3D-LaneNet:端到端三维多车道检测 
Sampling-Free Epistemic Uncertainty Estimation Using Approximated Variance Propagation基于近似方差传播无抽样认知不确定性估计 
Universal Adversarial Perturbation via Prior Driven Uncertainty Approximation基于先验驱动不确定近似普遍反对称扰动 
Understanding Deep Networks via Extremal Perturbations and Smooth Masks利用极值扰动光滑掩模理解深度网络 
Unsupervised Pre-Training of Image Features on Non-Curated Data非精确数据图像特征无监督预训练 
Learning Local Descriptors With a CDF-Based Dynamic Soft Margin基于CDF动态软边值实现局部描述子学习 
Bayes-Factor-VAE: Hierarchical Bayesian Deep Auto-Encoder Models for Factor DisentanglementBayes-Factor-VAE:用于因子分离分层Bayesian深度自编码模型 
Linearized Multi-Sampling for Differentiable Image Transformation基于线性化多重采样可微图像变换 
AdaTransform: Adaptive Data TransformationAdaTransform:自适应数据转换 
CARAFE: Content-Aware ReAssembly of FEaturesCARAFE:内容感知特征重组用于上采样的一种改进算法(如图2):分两步,首先训练出一个用于不同位置点乘的核(不同于双线性,不同位置的处理方式依赖于这个核);然后利用这个核来进行局部邻域的加权均值,从而实现不同位置,不同处理方式的上采样
AFD-Net: Aggregated Feature Difference Learning for Cross-Spectral Image Patch MatchingAFD-Net:用于跨光谱图像块匹配聚合特征差分学习 
Deep Joint-Semantics Reconstructing Hashing for Large-Scale Unsupervised Cross-Modal Retrieval面向大规模无监督跨模态检索深度联合语义重构哈希算法 
Unsupervised Neural Quantization for Compressed-Domain Similarity Search基于无监督神经量化压缩域相似性搜索 
Siamese Networks: The Tale of Two Manifolds孪生网络两个流形的故事 
Learning Combinatorial Embedding Networks for Deep Graph Matching用于深度图匹配组合嵌入网络学习 
Fashion Retrieval via Graph Reasoning Networks on a Similarity Pyramid基于相似金字塔图推理网络实现服装检索 
Wavelet Domain Style Transfer for an Effective Perception-Distortion Tradeoff in Single Image Super-Resolution单图像超分辨率基于小波域风格变换感知失真折衷 
Toward Real-World Single Image Super-Resolution: A New Benchmark and a New Model走向现实世界的单图像超分辨率:一种新的基准模型 
RankSRGAN: Generative Adversarial Networks With Ranker for Image Super-ResolutionRankSRGAN:基于RankerGAN实现图像超分辨率 
Progressive Fusion Video Super-Resolution Network via Exploiting Non-Local Spatio-Temporal Correlations利用非局部时空相关性渐进式融合实现视频超分辨率网络 
Deep SR-ITM: Joint Learning of Super-Resolution and Inverse Tone-Mapping for 4K UHD HDR Applications深度SR-ITM:4K超高清应用超分辨率逆色调映射联合学习 
Dynamic PET Image Reconstruction Using Nonnegative Matrix Factorization Incorporated With Deep Image Prior非负矩阵分解结合深度图像先验动态PET图像重建 
DSIC: Deep Stereo Image Compression深度立体图像压缩 
Variable Rate Deep Image Compression With a Conditional Autoencoder基于条件自动编码器变速率深度图像压缩 
Real Image Denoising With Feature Attention基于特征注意真实图像去噪 
Noise Flow: Noise Modeling With Conditional Normalizing Flows噪声流:使用条件规范化流噪声建模 
Bottleneck Potentials in Markov Random Fields马尔可夫随机场瓶颈势 
Seeing Motion in the Dark在黑暗中看运动 
SENSE: A Shared Encoder Network for Scene-Flow EstimationSENSE:用于场景流估计共享编码器网络 
Adversarial Feedback Loop对抗性反馈回路 
Dynamic-Net: Tuning the Objective Without Re-Training for Synthesis Tasks动态网无需重新训练即可调整目标实现综合任务 
AutoGAN: Neural Architecture Search for Generative Adversarial NetworksAutoGAN:生成性对抗网络神经结构搜索 
Co-Evolutionary Compression for Unpaired Image Translation基于协同进化压缩非成对图像翻译 
Self-Supervised Representation Learning From Multi-Domain Data多域数据的自监督表示学习 
Controlling Neural Networks via Energy Dissipation基于能量耗散神经网络控制 
Indices Matter: Learning to Index for Deep Image Matting索引的重要性:学习索引进行深度图像抠图 
LAP-Net: Level-Aware Progressive Network for Image DehazingLAP-Net:基于层级感知递进网络图像去雾 
Attention Augmented Convolutional Networks注意力增强卷积网络 
MetaPruning: Meta Learning for Automatic Neural Network Channel Pruning元剪枝:神经网络通道自动剪枝元学习 
Accelerate CNN via Recursive Bayesian Pruning通过递归贝叶斯剪枝实现加速CNN 
HBONet: Harmonious Bottleneck on Two Orthogonal DimensionsHBONet:两个正交维度上的和谐瓶颈 
O2U-Net: A Simple Noisy Label Detection Approach for Deep Neural NetworksO2U-Net:一种简单的深度神经网络中噪声标签检测方法 
Continual Learning by Asymmetric Loss Approximation With Single-Side Overestimation基于单侧高估非对称损失逼近实现连续学习 
Label-PEnet: Sequential Label Propagation and Enhancement Networks for Weakly Supervised Instance SegmentationLabel-PEnet:基于序列标签传播增强网络弱监督实例分割 
LIP: Local Importance-Based PoolingLIP:局部基于重要性池化 
Global Feature Guided Local Pooling全局功能引导局部池化 
Conditional Coupled Generative Adversarial Networks for Zero-Shot Domain Adaptation基于条件耦合GAN零镜头域自适应 
Adversarial Defense by Restricting the Hidden Space of Deep Neural Networks通过限制深层神经网络隐藏空间实现对抗防御 
Hyperpixel Flow: Semantic Correspondence With Multi-Layer Neural Features超像素流:基于多层神经特征语义对应 
Information Entropy Based Feature Pooling for Convolutional Neural Networks基于信息熵卷积神经网络特征池 
Patchwork: A Patch-Wise Attention Network for Efficient Object Detection and Segmentation in Video StreamsPatchWork:一种用于视频流中有效目标检测分割补丁式注意力网络 
AttentionRNN: A Structured Spatial Attention MechanismAttentionRNN:一种结构化空间注意机制像RNN一样的Attention,即在估计Attention Mask时,每个点都依赖于前面已估计出的点(传统的方式是,每个点独立估计)
Drop an Octave: Reducing Spatial Redundancy in Convolutional Neural Networks With Octave Convolution降八度:用八度卷积减少卷积神经网络的空间冗余 
Domain Intersection and Domain Difference域交域差 
Learned Video Compression学习视频压缩 
Local Relation Networks for Image Recognition基于局部关系网络图像识别 
DiscoNet: Shapes Learning on Disconnected Manifolds for 3D EditingDiscoNect:断开流形上的形状学习实现三维编辑 
Deep Residual Learning in the JPEG Transform DomainJPEG变换域深度残差学习 
Approximated Bilinear Modules for Temporal Modeling基于近似双线性模型时域建模 
Customizing Student Networks From Heterogeneous Teachers via Adaptive Knowledge Amalgamation自适应知识融合实现异构教师网络定制学生网络 
Data-Free Learning of Student Networks学生网络无数据学习 
Deep Closest Point: Learning Representations for Point Cloud Registration深度最近点:基于表示学习点云配准 
Orientation-Aware Semantic Segmentation on Icosahedron Spheres二十面体球面上方向感知语义分割全方向(omnidirectional)图像的语义分割
Differentiable Learning-to-Group Channels via Groupable Convolutional Neural Networks基于可分组卷积神经网络信道群可微学习 
HarDNet: A Low Memory Traffic NetworkHarDNet:一个低内存交通网络 
Dynamic Multi-Scale Filters for Semantic Segmentation用于语义分割动态多尺度滤波器如图2,网络中添加多个个基于自适应池化学习出来的滤波器
Online Model Distillation for Efficient Video Inference基于在线模型蒸馏有效视频推理 
Rethinking Zero-Shot Learning: A Conditional Visual Classification Perspective条件视觉分类的角度反思零镜头学习 
Task-Driven Modular Networks for Zero-Shot Compositional Learning基于任务驱动模块化网络零镜头组合学习 
Transductive Episodic-Wise Adaptive Metric for Few-Shot Learning基于转导不定自适应度量少数镜头学习 
Deep Multiple-Attribute-Perceived Network for Real-World Texture Recognition用于真实纹理识别深度多属性感知网络 
RGB-Infrared Cross-Modality Person Re-Identification via Joint Pixel and Feature Alignment基于联合像素特征对齐RGB-红外交叉模态人再识别 
EvalNorm: Estimating Batch Normalization Statistics for EvaluationEvalNorm:估计用于评估的批处理规范化(BN)统计信息 
Beyond Human Parts: Dual Part-Aligned Representations for Person Re-Identification超越人的部分:基于双部分对齐表示的人再识别 
Person Search by Text Attribute Query As Zero-Shot Learning基于作为零镜头学习文本属性查询人搜索算法 
Semantic-Aware Knowledge Preservation for Zero-Shot Sketch-Based Image Retrieval语义感知知识保存实现零镜头基于草图的图像检索 
Active Learning for Deep Detection Neural Networks主动学习实现深度检测神经网络 
One-Shot Neural Architecture Search via Self-Evaluated Template Network基于自评估模板网络一次性神经网络结构搜索 
Batch DropBlock Network for Person Re-Identification and Beyond用于人再识别及其他的批处理DropBlock网络 
Omni-Scale Feature Learning for Person Re-Identification全尺度特征学习用于人再识别 
Be Your Own Teacher: Improve the Performance of Convolutional Neural Networks via Self Distillation做自己的老师:通过自蒸馏提高卷积神经网络的性能 
Diversity With Cooperation: Ensemble Methods for Few-Shot Classification合作分集:用于少镜头分类的集成方法 
Enhancing 2D Representation via Adjacent Views for 3D Shape Retrieval基于邻接视图二维图形增强实现三维形状检索 
Adversarial Fine-Grained Composition Learning for Unseen Attribute-Object Recognition对抗性细粒度合成学习不可见属性-对象识别中的应用 
Auto-ReID: Searching for a Part-Aware ConvNet for Person Re-IdentificationAuto-ReID:搜索局部感知ConvNet实现人重识别 
Second-Order Non-Local Attention Networks for Person Re-Identification二阶非局部注意网络用于人再识别 
Fast Computation of Content-Sensitive Superpixels and Supervoxels Using Q-DistancesQ-距离快速计算内容敏感超像素超体素 
Progressive-X: Efficient, Anytime, Multi-Model Fitting AlgorithmProgressive-X:高效、随时、多模型拟合算法 
Structured Modeling of Joint Deep Feature and Prediction Refinement for Salient Object Detection联合深度特征预测细化结构化建模实现显著目标检测 
Selectivity or Invariance: Boundary-Aware Salient Object Detection选择性或不变性:边界感知显著目标检测 
Online Unsupervised Learning of the 3D Kinematic Structure of Arbitrary Rigid Bodies任意刚体三维运动结构在线无监督学习 
Few-Shot Generalization for Single-Image 3D Reconstruction via Priors利用少镜头泛化实现基于先验的单幅图像三维重建 
Digging Into Self-Supervised Monocular Depth Estimation自监督单目深度估计方法的研究 
Learning Object-Specific Distance From a Monocular Image从单目图像学习特定对象距离 
Unsupervised 3D Reconstruction Networks无监督三维重建网络 
3D Point Cloud Generative Adversarial Network Based on Tree Structured Graph Convolutions基于树结构图卷积三维点云GAN 
Visualization of Convolutional Neural Networks for Monocular Depth Estimation卷积神经网络可视化单目深度估计中的应用 
Co-Separating Sounds of Visual Objects视觉对象的共分离声音 
BMN: Boundary-Matching Network for Temporal Action Proposal GenerationBMN:基于边界匹配网络时间行为建议生成 
Weakly Supervised Temporal Action Localization Through Contrast Based Evaluation Networks基于对比度评价网络弱监督时间行为定位 
Progressive Sparse Local Attention for Video Object Detection基于渐进稀疏局部注意视频目标检测 
Reasoning About Human-Object Interactions Through Dual Attention Networks基于双注意网络人机交互推理 
DMM-Net: Differentiable Mask-Matching Network for Video Object SegmentationDMM-Net:用于视频对象分割可微掩模匹配网络 
Asymmetric Cross-Guided Attention Network for Actor and Action Video Segmentation From Natural Language Query非对称交叉引导注意网络实现自然语言查询中角色动作视频分割 
AGSS-VOS: Attention Guided Single-Shot Video Object SegmentationAGSS-VOS:注意力引导单镜头视频对象分割 
Global-Local Temporal Representations for Video Person Re-Identification基于全局-局部时间表示视频人再识别 
AdvIT: Adversarial Frames Identifier Based on Temporal Consistency in VideosADvIT:基于时间一致性视频对抗帧标识符 
RANet: Ranking Attention Network for Fast Video Object SegmentationRANet:用于视频对象快速分割排序注意网络 
Spatial-Temporal Relation Networks for Multi-Object Tracking用于多目标跟踪时空关系网络 
Bridging the Gap Between Detection and Tracking: A Unified Approach缩小检测跟踪之间的差距:一种统一的方法 
Learning the Model Update for Siamese Trackers学习孪生跟踪器模型更新 
Fast-deepKCF Without Boundary Effect无边界效应的快速深度KCF 
Program-Guided Image Manipulators程序引导图像操纵器 
Calibration of Axial Fisheye Cameras Through Generic Virtual Central Models通用虚拟中心模型鱼眼相机的标定 
Micro-Baseline Structured Light微基线结构光 
l-Net: Reconstruct Hyperspectral Images From a Snapshot Measurementl-Net:从快照测量重建高光谱图像 
Deep Depth From Aberration Map像差图深度 
A Dataset of Multi-Illumination Images in the Wild野外多光照图像数据集 
Monocular Neural Image Based Rendering With Continuous View Control利用连续视图控制实现基于单目神经图像展示 
Multi-View Image Fusion多视点图像融合 
Enhancing Low Light Videos by Exploring High Sensitivity Camera Noise利用高灵敏度相机噪声实现微光视频增强 
Deep Restoration of Vintage Photographs From Scanned Halftone Prints扫描的半色调照片深度复原复古照片 
Context-Aware Image Matting for Simultaneous Foreground and Alpha Estimation上下文感知图像抠图实现同时进行前景和α估计 
CFSNet: Toward a Controllable Feature Space for Image RestorationCFSNet:基于可控特征空间图像复原 
Deep Blind Hyperspectral Image Fusion深度盲高光谱图像融合 
Fully Convolutional Pixel Adaptive Image Denoiser全卷积像素自适应图像去噪 
Coherent Semantic Attention for Image Inpainting基于连贯语义注意图像修补 
Embedded Block Residual Network: A Recursive Restoration Model for Single-Image Super-Resolution嵌入块残差网络:一种单图像超分辨率递归恢复模型 
Fast Image Restoration With Multi-Bin Trainable Linear Units基于Multi-Bin可训练线性单元快速图像复原 
Counting With Focus for Free免费焦点计数 
SynDeMo: Synergistic Deep Feature Alignment for Joint Learning of Depth and Ego-MotionSynDeMo:基于协同深度特征对齐深度自我运动联合学习 
Diverse Image Synthesis From Semantic Layouts via Conditional IMLE基于条件IMLE的语义布局多样性图像合成 
Towards Bridging Semantic Gap to Improve Semantic Segmentation通过桥接语义鸿沟实现语义分割改进文章关注不同尺度特征的融合问题,在图6的网络结构中,使用了图4的三个模块,主要从多尺度融合和边缘感知两个方向,提升语义分割的效果
Generating Diverse and Descriptive Image Captions Using Visual Paraphrases使用视觉释义生成多样的描述性图片标注 
Learning to Collocate Neural Modules for Image Captioning基于神经模块配置学习图像标注 
Sequential Latent Spaces for Modeling the Intention During Diverse Image Captioning序列潜空间多样图像标注中的意图建模 
Why Does a Visual Question Have Different Answers?为什么视觉问题不同的答案 
G3raphGround: Graph-Based Language GroundingG3raphGround:基于图形语言Grounding 
Scene Text Visual Question Answering场景文本可视化问答 
Unsupervised Collaborative Learning of Keyframe Detection and Visual Odometry Towards Monocular Deep SLAM关键帧检测视觉里程测量无监督协同学习实现单目深度SLAM 
MVSCRF: Learning Multi-View Stereo With Conditional Random FieldsMVSCRF:基于条件随机场多视图立体学习 
Neural-Guided RANSAC: Learning Where to Sample Model Hypotheses神经引导的RANSAC模型假设采样位置学习 
Efficient Learning on Point Clouds With Basis Point Sets基于基础点集点云高效学习 
Cross View Fusion for 3D Human Pose Estimation基于交叉视图融合三维人体姿态估计 
Shape-Aware Human Pose and Shape Reconstruction Using Multi-View Images基于多视点图像形状感知人体姿态形状重建 
Monocular Piecewise Depth Estimation in Dynamic Scenes by Exploiting Superpixel Relations基于超像素关系动态场景单目分段深度估计 
Is This the Right Place? Geometric-Semantic Pose Verification for Indoor Visual Localization这是对的地方吗?基于几何-语义位姿验证室内视觉定位 
DeepPruner: Learning Efficient Stereo Matching via Differentiable PatchMatchDeepPruner:通过可微Patch匹配实现有效的立体匹配学习1. 利用RNN的结构,描述PatchMatch
2. 利用可微的PatchMatch,缩小每个像素视差的搜索范围(传统的方法是所有视差可能性,而文中每个像素考虑的是部分视差,即Confidence Range,大约是全部视差范围的1/10)
Convolutional Sequence Generation for Skeleton-Based Action Synthesis利用卷积序列生成实现基于骨架的动作合成 
Onion-Peel Networks for Deep Video CompletionOnion-Peel网络用于深度视频补全 
Copy-and-Paste Networks for Deep Video Inpainting基于复制-粘贴网络深度视频修补 
Content and Style Disentanglement for Artistic Style Transfer基于内容与风格解构艺术风格转换 
Image2StyleGAN: How to Embed Images Into the StyleGAN Latent Space?Image2StyleGAN:如何图像嵌入StyleGAN潜在空间 
Controllable Artistic Text Style Transfer via Shape-Matching GAN基于形状-匹配GAN可控艺术文本风格转换 
Understanding Generalized Whitening and Coloring Transform for Universal Style Transfer广义白化着色变换通用风格转换中的应用 
Learning Implicit Generative Models by Matching Perceptual Features基于感知特征匹配隐生成模型学习 
Free-Form Image Inpainting With Gated Convolution基于门控卷积自由形式图像补全 
FiNet: Compatible and Diverse Fashion Image InpaintingFiNet:兼容的和多样时尚形象修复 
InGAN: Capturing and Retargeting the "DNA" of a Natural ImageInGAN:捕捉重新定位自然图像的“DNA” 
Seeing What a GAN Cannot Generate看一个GAN不能产生什么 
COCO-GAN: Generation by Parts via Conditional CoordinatingCOCO-GAN:基于条件配位分块生成 
Neural Turtle Graphics for Modeling City Road Layouts基于神经海龟图形建模城市道路规划 
Texture Fields: Learning Texture Representations in Function Space纹理场:在函数空间学习纹理表示 
PointFlow: 3D Point Cloud Generation With Continuous Normalizing FlowsPointFlow:基于连续规格化流三维点云生成 
Meta-Sim: Learning to Generate Synthetic DatasetsMeta-Sim:学习生成合成数据集 
Specifying Object Attributes and Relations in Interactive Scene Generation交互式场景生成指定对象属性关系 
SinGAN: Learning a Generative Model From a Single Natural ImageSinGAN:从单一自然图像学习生成模型 
VaTeX: A Large-Scale, High-Quality Multilingual Dataset for Video-and-Language ResearchVaTex:一个用于视频和语言研究大规模、高质量的多语言数据集 
A Graph-Based Framework to Bridge Movies and Synopses一种基于的框架实现电影与剧情的桥接 
From Strings to Things: Knowledge-Enabled VQA Model That Can Read and Reason从字符串到事物:可以读取和推理支持知识的VQA模型 
Counterfactual Critic Multi-Agent Training for Scene Graph Generation用于场景图生成反事实批评家多智能体训练 
Robust Change Captioning强大的更改字幕 
Attention on Attention for Image Captioning  
Dynamic Graph Attention for Referring Expression Comprehension动态图形注意力指称表达理解中的应用 
Visual Semantic Reasoning for Image-Text Matching基于视觉语义推理图-文匹配 
Phrase Localization Without Paired Training Examples无配对训练实例短语定位 
Learning to Assemble Neural Module Tree Networks for Visual Grounding基于神经模块树网络学习视觉Grounding 
A Fast and Accurate One-Stage Approach to Visual Grounding一种快速准确的视觉Grounding方法 
Zero-Shot Grounding of Objects From Natural Language Queries基于自然语言查询对象的零镜头Grounding 
Towards Unconstrained End-to-End Text Spotting朝向无约束端到端文本定位 
What Is Wrong With Scene Text Recognition Model Comparisons? Dataset and Model Analysis场景文本识别模型比较有什么问题?数据集模型分析 
Sparse and Imperceivable Adversarial Attacks稀疏而难以想象的对抗性攻击 
Enhancing Adversarial Example Transferability With an Intermediate Level Attack使用中级攻击增强对手示例可转移性 
Implicit Surface Representations As Layers in Neural Networks神经网络层中隐式曲面表示 
A Tour of Convolutional Networks Guided by Linear Interpreters线性解释引导卷积网络之旅 
Small Steps and Giant Leaps: Minimal Newton Solvers for Deep Learning小步大步深度学习最小牛顿解 
Semantic Adversarial Attacks: Parametric Transformations That Fool Deep Classifiers语义对抗攻击:通过参数转换愚弄深度分类器 
Hilbert-Based Generative Defense for Adversarial Examples基于希尔伯特生成性防御实现对抗例子 
On the Efficacy of Knowledge Distillation知识蒸馏功效 
Sym-Parameterized Dynamic Inference for Mixed-Domain Image Translation混合域图像翻译Sym参数化动态推理 
Better and Faster: Exponential Loss for Image Patch Matching更快更好:图像块匹配指数损失 
Physical Adversarial Textures That Fool Visual Object Tracking物理对抗纹理欺骗视觉对象跟踪 
Wasserstein GAN With Quadratic Transport Cost基于二次传输代价Wasserstein GAN 
Scalable Verified Training for Provably Robust Image Classification基于可扩展验证训练可证明鲁棒图像分类 
Differentiable Soft Quantization: Bridging Full-Precision and Low-Bit Neural Networks可微软量化全精度低比特神经网络桥接 
The LogBarrier Adversarial Attack: Making Effective Use of Decision Boundary InformationLogBarrier对抗攻击决策边界信息的有效利用 
Proximal Mean-Field for Neural Network Quantization基于近场平均场神经网络量化 
Improving Adversarial Robustness via Guided Complement Entropy利用引导互补熵提高对抗稳健性 
A Geometry-Inspired Decision-Based Attack基于几何启发决策攻击 
Universal Perturbation Attack Against Image Retrieval图像检索中的普遍扰动攻击 
Bayesian Optimized 1-Bit CNNs贝叶斯优化的1-BitCNNs 
Rethinking ImageNet Pre-TrainingImageNet预训练再思考 
Defending Against Universal Perturbations With Shared Adversarial Training基于共同对抗性训练普遍干扰防御 
Adaptive Activation Thresholding: Dynamic Routing Type Behavior for Interpretability in Convolutional Neural Networks自适应激活阈值:基于动态路由类型行为卷积神经网络可解释性 
XRAI: Better Attributions Through RegionsXRAI:通过区域获得更好的属性 
Guessing Smart: Biased Sampling for Efficient Black-Box Adversarial Attacks猜测智能:基于有偏抽样高效黑盒对抗攻击 
Mask-Guided Attention Network for Occluded Pedestrian Detection基于面罩引导注意网络遮挡行人检测 
Spectral Feature Transformation for Person Re-Identification基于谱特征变换人再识别 
Permutation-Invariant Feature Restructuring for Correlation-Aware Image Set-Based Recognition置换不变特征重构实现基于相关感知图像集的图像识别 
Improving Pedestrian Attribute Recognition With Weakly-Supervised Multi-Scale Attribute-Specific Localization基于弱监督多尺度属性特定定位行人属性识别 
Correlation Congruence for Knowledge Distillation基于相关同余知识蒸馏 
Dynamic Curriculum Learning for Imbalanced Data Classification基于动态课程学习不平衡数据分类 
Video Face Clustering With Unknown Number of Clusters未知簇数的视频人脸聚类 
Targeted Mismatch Adversarial Attack: Query With a Flower to Retrieve the Tower目标不匹配对抗攻击查询以检索 
Fashion++: Minimal Edits for Outfit ImprovementFashion++:以最小编辑实现服装改进 
Semi-Supervised Pedestrian Instance Synthesis and Detection With Mutual Reinforcement基于互增强半监督行人实例综合检测 
SILCO: Show a Few Images, Localize the Common ObjectSILCO:显示一些图像定位公共对象 
A Deep Step Pattern Representation for Multimodal Retinal Image Registration多模视网膜图像配准深度阶跃模式表示 
Deep Graphical Feature Learning for the Feature Matching Problem深度图形特征学习解决特征匹配问题 
Minimum Delay Object Detection From Video视频的最小延迟目标检测 
Learning With Average Precision: Training Image Retrieval With a Listwise Loss平均精度学习:基于列表损失图像检索训练 
Learning to Find Common Objects Across Few Image Collections学习在少数图像集合查找公共对象 
Weakly Aligned Cross-Modal Learning for Multispectral Pedestrian Detection基于弱对齐交叉模式学习多光谱行人检测 
Deep Self-Learning From Noisy Labels嘈杂的标签中深度自我学习 
DSConv: Efficient Convolution OperatorDSConv:高效卷积算子 
Once a MAN: Towards Multi-Target Attack via Learning Multi-Target Adversarial Network Once一人一次:通过一次学习多目标对抗网络实现多目标攻击 
Explicit Shape Encoding for Real-Time Instance Segmentation基于显式形状编码实时实例分割 
IMP: Instance Mask Projection for High Accuracy Semantic Segmentation of ThingsIMP:用于高精度语义分割实例掩码投影 
Video Instance Segmentation视频实例分割 
Attention Bridging Network for Knowledge Transfer基于注意力桥接网络知识转移 
Self-Supervised Difference Detection for Weakly-Supervised Semantic Segmentation基于自监督差分检测弱监督语义分割 
SPGNet: Semantic Prediction Guidance for Scene ParsingSPGNet:基于语义预测指导场景分析 
Gated-SCNN: Gated Shape CNNs for Semantic Segmentation门控SCNN:用于语义分割门控形状CNN 
DensePoint: Learning Densely Contextual Representation for Efficient Point Cloud ProcessingDensePoint:基于密集上下文表示学习的高效点云处理 
AMP: Adaptive Masked Proxies for Few-Shot SegmentationAMP:基于自适应掩蔽代理少镜头分割 
Universal Semi-Supervised Semantic Segmentation通用半监督语义分割 
Accelerate Learning of Deep Hashing With Gradient Attention利用梯度注意力加速深度散列学习 
SVD: A Large-Scale Short Video Dataset for Near-Duplicate Video RetrievalSVD:一种用于近重复视频检索大规模短视频数据集 
Block Annotation: Better Image Annotation With Sub-Image Decomposition块注释:使用子图像分解更好的图像注释 
Probabilistic Deep Ordinal Regression Based on Gaussian Processes基于高斯过程概率深度序数回归 
Balanced Datasets Are Not Enough: Estimating and Mitigating Gender Bias in Deep Image Representations平衡的数据集是不够的:估计减轻深度图像表现中的性别偏见 
Teacher Guided Architecture Search教师指导架构搜索 
FACSIMILE: Fast and Accurate Scans From an Image in Less Than a SecondFACSIMILE:在不到一秒钟的时间内快速准确地扫描图像 
Delving Deep Into Hybrid Annotations for 3D Human Recovery in the Wild深入研究混合标注野外三维人体复原中的应用 
Human Mesh Recovery From Monocular Images via a Skeleton-Disentangled Representation基于骨架分离表示单目图像人体网格恢复 
Three-D Safari: Learning to Estimate Zebra Pose, Shape, and Texture From Images "In the Wild"三维漫游:学习从“野外”图像估计斑马姿势形状纹理 
Object-Driven Multi-Layer Scene Decomposition From a Single Image基于单个图像的对象驱动多层场景分解 
Occupancy Flow: 4D Reconstruction by Learning Particle Dynamics占用流:基于粒子动力学4d重建 
Joint Monocular 3D Vehicle Detection and Tracking单目三维车辆联合检测跟踪 
Fingerspelling Recognition in the Wild With Iterative Visual Attention基于迭代视觉注意力野外手指拼写识别 
PointAE: Point Auto-Encoder for 3D Statistical Shape and Texture ModellingPointAE:用于三维统计形状纹理建模点自动编码器 
Multi-Garment Net: Learning to Dress 3D People From Images多服装网:从图像中学习三维人体着装 
Skeleton-Aware 3D Human Shape Reconstruction From Point Clouds基于点云骨骼感知三维人体形状重建 
AMASS: Archive of Motion Capture As Surface ShapesAMASS:作为表面形状运动捕捉存档 
Person-in-WiFi: Fine-Grained Person Perception Using WiFiWIFI中的人:使用WIFI细粒度的人感知 
FAB: A Robust Facial Landmark Detection Framework for Motion-Blurred VideosFAB:一种鲁棒的运动模糊视频人脸地标检测框架 
Attentional Feature-Pair Relation Networks for Accurate Face Recognition基于注意力特征对关系网络精确人脸识别 
Action Recognition With Spatial-Temporal Discriminative Filter Banks基于时空判别滤波器组动作识别 
EPIC-Fusion: Audio-Visual Temporal Binding for Egocentric Action RecognitionEPIC融合:用于自我中心行为识别视听时间绑定 
Weakly-Supervised Action Localization With Background Modeling基于背景建模弱监督动作定位 
Grouped Spatial-Temporal Aggregation for Efficient Action Recognition基于分组时空聚合动作识别 
Temporal Structure Mining for Weakly Supervised Action Detection弱监督动作检测时间结构挖掘 
Temporal Recurrent Networks for Online Action Detection用于在线动作检测时间递归网络 
StartNet: Online Detection of Action Start in Untrimmed VideosStartNet:未剪辑视频动作开始的在线检测 
Video Classification With Channel-Separated Convolutional Networks基于通道分离卷积网络视频分类 
Predicting the Future: A Jointly Learnt Model for Action Anticipation预测未来:一个基于共同学习行动预测模型 
Human-Aware Motion Deblurring人体感知运动去模糊 
Fast Video Object Segmentation via Dynamic Targeting Network基于动态目标网络视频对象快速分割 
Solving Vision Problems via Filtering通过滤波解决视觉问题 
GAN-Based Projector for Faster Recovery With Convergence Guarantees in Linear Inverse Problems线性反问题基于GAN投影实现具有收敛保证的更快恢复 
Scoot: A Perceptual Metric for Facial SketchesScoot:基于感知测度面部草图 
Learning Filter Basis for Convolutional Neural Network Compression基于滤波基学习卷积神经网络压缩 
End-to-End Learning of Representations for Asynchronous Event-Based Data端到端表示学习实现异步基于事件的数据 
ERL-Net: Entangled Representation Learning for Single Image De-RainingERL网:基于纠缠表示学习单图像去雨 
Perceptual Deep Depth Super-Resolution感知深度超分辨率 
3D Scene Graph: A Structure for Unified Semantics, 3D Space, and Camera三维场景图:用于统一语义三维空间相机结构 
Floorplan-Jigsaw: Jointly Estimating Scene Layout and Aligning Partial Scans平面拼图:联合估计场景布局对齐部分扫描 
Enforcing Geometric Constraints of Virtual Normal for Depth Prediction基于虚拟法向几何约束深度预测 
Deep Contextual Attention for Human-Object Interaction Detection基于深度上下文注意人-对象交互检测 
Learning Compositional Neural Information Fusion for Human Parsing用于人类分析合成神经信息融合学习 
Attentional Neural Fields for Crowd Counting人群计数注意神经场 
Understanding Human Gaze Communication by Spatio-Temporal Graph Reasoning时空图推理理解人的凝视交流 
Controllable Attention for Structured Layered Video Decomposition基于可控注意结构化分层视频分解 
GANalyze: Toward Visual Definitions of Cognitive Image Properties认知图像属性视觉定义 
Saliency-Guided Attention Network for Image-Sentence Matching显著性引导注意力网络图像-句子匹配中的应用 
CAMP: Cross-Modal Adaptive Message Passing for Text-Image RetrievalCAMP:用于文本-图像检索跨模式自适应消息传递 
ACMM: Aligned Cross-Modal Memory for Few-Shot Image and Sentence MatchingACMM:用于少镜头图像句子匹配对齐跨模态存储器 
Creativity Inspired Zero-Shot Learning创意激发零镜头学习 
Generating Easy-to-Understand Referring Expressions for Target Identifications目标识别生成易于理解的指代表达 
Language-Agnostic Visual-Semantic Embeddings语言不可知视觉语义嵌入 
Adversarial Representation Learning for Text-to-Image Matching文本-图像匹配中的对抗表示学习 
Multi-Modality Latent Interaction Network for Visual Question Answering视觉问答多模态潜在交互网络 
Key.Net: Keypoint Detection by Handcrafted and Learned CNN FiltersKey.Net:基于手工特征CNN过滤器学习关键点检测 
Learning Two-View Correspondences and Geometry Using Order-Aware Network基于顺序感知网络两视图对应几何学习 
Learning Meshes for Dense Visual SLAM稠密视觉SLAM学习网格 
EM-Fusion: Dynamic Object-Level SLAM With Probabilistic Data AssociationEM融合:基于概率数据关联动态对象级SLAM 
ClusterSLAM: A SLAM Backend for Simultaneous Rigid Body Clustering and Motion EstimationClusterSLAM同时进行刚体聚类运动估计SLAM后端 
Efficient and Robust Registration on the 3D Special Euclidean Group三维特殊欧氏群高效鲁棒配准 
Algebraic Characterization of Essential Matrices and Their Averaging in Multiview Settings多视图环境本质矩阵代数特征其平均 
Liquid Warping GAN: A Unified Framework for Human Motion Imitation, Appearance Transfer and Novel View Synthesis液体翘曲GAN:一个统一的人体运动模拟外观传递新视角合成框架 
RelGAN: Multi-Domain Image-to-Image Translation via Relative AttributesRelGAN:基于相对属性多域图像-图像转换 
Attribute-Driven Spontaneous Motion in Unpaired Image Translation非成对图像翻译中的属性驱动自发运动 
Everybody Dance Now现在大家都跳舞 
Multimodal Style Transfer via Graph Cuts基于图割多模态转移 
A Closed-Form Solution to Universal Style Transfer通用样式转换的一种闭式解法 
Progressive Reconstruction of Visual Structure for Image Inpainting图像修补视觉结构渐进重建 
Variational Adversarial Active Learning变分对抗性主动学习主动学习:让学习算法主动地提出要对哪些数据进行标注
Confidence Regularized Self-Training基于自信心正则化自训练 
Anchor Loss: Modulating Loss Scale Based on Prediction Difficulty锚损失:基于预测难度调整损失尺度 
Local Aggregation for Unsupervised Learning of Visual Embeddings基于局部聚集无监督视觉嵌入学习 
PR Product: A Substitute for Inner Product in Neural NetworksPR乘积神经网络内积一种代用品 
CutMix: Regularization Strategy to Train Strong Classifiers With Localizable FeaturesCutMix:训练具有局部特征的强分类器正则化策略 
Towards Interpretable Object Detection by Unfolding Latent Structures基于潜在结构展开可解释目标检测 
Scaling Object Detection by Transferring Classification Weights基于分类权重转移分级目标检测 
Scale-Aware Trident Networks for Object Detection基于尺度感知的Trident网络实现目标检测 
Object-Aware Instance Labeling for Weakly Supervised Object Detection基于目标感知实例标记弱监督目标检测 
Generative Modeling for Small-Data Object Detection小数据目标检测生成模型 
Transductive Learning for Zero-Shot Object Detection基于导纳学习零镜头目标检测 
Self-Training and Adversarial Background Regularization for Unsupervised Domain Adaptive One-Stage Object Detection基于自训练对抗背景正则化无监督域自适应单阶段目标检测 
Memory-Based Neighbourhood Embedding for Visual Recognition基于记忆的邻域嵌入实现视觉识别 
Self-Similarity Grouping: A Simple Unsupervised Cross Domain Adaptation Approach for Person Re-Identification自相似分组:一种简单的无监督跨域自适应方法 
Deep Reinforcement Active Learning for Human-in-the-Loop Person Re-Identification基于深度强化主动学习回路中人的人再识别 
A Dual-Path Model With Adaptive Attention for Vehicle Re-Identification一种具有自适应注意的双路径模型实现车辆重识别 
Bayesian Loss for Crowd Count Estimation With Point Supervision贝叶斯损失用于基于点监督的人群计数 
Learning Spatial Awareness to Improve Crowd Counting空间感知学习提高人群计数 
GradNet: Gradient-Guided Network for Visual Object TrackingGradNet:基于梯度引导网络视觉目标跟踪 
FAMNet: Joint Learning of Feature, Affinity and Multi-Dimensional Assignment for Online Multiple Object TrackingFAMNet:基于特征亲和力多维分配联合学习在线多目标跟踪 
Learning Discriminative Model Prediction for Tracking基于判别模型预测学习跟踪 
DynamoNet: Dynamic Action and Motion Network动态动作运动网络 
SlowFast Networks for Video Recognition用于视频识别SlowFast网络 
Generative Multi-View Human Action Recognition生成性多视角人类行为识别 
Multi-Agent Reinforcement Learning Based Frame Sampling for Effective Untrimmed Video Recognition基于多智能体增强学习帧采样实现有效的未经修剪视频识别 
SCSampler: Sampling Salient Clips From Video for Efficient Action RecognitionSCSampler:从视频中抽取显著片段以实现高效的动作识别 
Weakly Supervised Energy-Based Learning for Action Segmentation弱监督基于能量的学习实现动作分割 
What Would You Expect? Anticipating Egocentric Actions With Rolling-Unrolling LSTMs and Modality Attention你期望什么?以滚动-展开的LSTMs情态注意预测自我中心行为 
PIE: A Large-Scale Dataset and Models for Pedestrian Intention Estimation and Trajectory PredictionPIE:用于行人意图估计轨迹预测大规模数据集和模型 
STGAT: Modeling Spatial-Temporal Interactions for Human Trajectory PredictionSTGAT:用于人类轨迹预测时-空交互建模 
Learning Motion in Feature Space: Locally-Consistent Deformable Convolution Networks for Fine-Grained Action Detection特征空间中的运动学习:基于局部一致可变形卷积网络细粒度动作检测 
Dual Attention Matching for Audio-Visual Event Localization基于双注意匹配视-听事件定位 
Uncertainty-Aware Audiovisual Activity Recognition Using Deep Bayesian Variational Inference基于深度贝叶斯变分推理不确定性感知的视-听活动识别 
Non-Local Recurrent Neural Memory for Supervised Sequence Modeling基于非局部递归神经记忆监督序列建模 
Temporal Attentive Alignment for Large-Scale Video Domain Adaptation基于时间注意力对齐大规模视频域自适应 
Action Assessment by Joint Relation Graphs基于联合关系图行动评估 
Unsupervised Procedure Learning via Joint Dynamic Summarization基于联合动态摘要无监督过程学习 
ViSiL: Fine-Grained Spatio-Temporal Video Similarity LearningViSiL:细粒度时-空视频相似度学习 
Unsupervised Learning of Landmarks by Descriptor Vector Exchange基于描述向量交换无监督地标学习 
Learning Compositional Representations for Few-Shot Recognition基于合成表示学习少镜头识别 
Spectral Regularization for Combating Mode Collapse in GANs基于谱正则化GANs抗模式崩溃 
Scaling and Benchmarking Self-Supervised Visual Representation Learning自监督视觉表示学习标度标杆 
Learning an Effective Equivariant 3D Descriptor Without Supervision无监督学习一种有效的等变三维描述子 
KPConv: Flexible and Deformable Convolution for Point CloudsKPConv:用于点云柔性可变形卷积 
Neural Inter-Frame Compression for Video Coding基于神经帧间压缩视频编码 
Task2Vec: Task Embedding for Meta-LearningTask2Vec:基于任务嵌入元学习 
Deep Clustering by Gaussian Mixture Variational Autoencoders With Graph Embedding图嵌入实现基于高斯混合变分自编码深度聚类 
SoftTriple Loss: Deep Metric Learning Without Triplet Sampling软三元损失无三元抽样的深度度量学习 
A Weakly Supervised Fine Label Classifier Enhanced by Coarse Supervision一种基于粗监督弱监督精细标记分类器 
Gaussian Affinity for Max-Margin Class Imbalanced Learning基于高斯亲合性最大边缘类非平衡学习 
AttPool: Towards Hierarchical Feature Representation in Graph Convolutional Networks via Attention MechanismAttPool:基于注意力机制图形卷积网络层次特征表示 
Deep Metric Learning With Tuplet Margin Loss具有三元边缘损失深度度量学习 
Normalized Wasserstein for Mixture Distributions With Applications in Adversarial Learning and Domain Adaptation基于标准化Wasserstein的混合分布对抗学习域自适应中的应用 
Fast and Practical Neural Architecture Search快速实用的神经网络架构搜索 
Symmetric Graph Convolutional Autoencoder for Unsupervised Graph Representation Learning基于对称图卷积自动编码器无监督图表示学习 
Deep Elastic Networks With Model Selection for Multi-Task Learning基于模型选择的深度弹性网络实现多任务学习 
Metric Learning With HORDE: High-Order Regularizer for Deep Embeddings基于HORDE度量学习深度嵌入高阶正则化 
Adversarial Learning With Margin-Based Triplet Embedding Regularization基于边缘的三元嵌入正则化实现对抗学习 
Simultaneous Multi-View Instance Detection With Learned Geometric Soft-Constraints基于学习几何软约束多视图同时实例检测 
CenterNet: Keypoint Triplets for Object DetectionCenterNet:基于关键点三元组对象检测 
Online Hyper-Parameter Learning for Auto-Augmentation Strategy基于在线超参数学习自增强策略 
DANet: Divergent Activation for Weakly Supervised Object LocalizationDANet:基于发散激活弱监督目标定位 
Selective Sparse Sampling for Fine-Grained Image Recognition基于选择性稀疏采样细粒度图像识别 
Dynamic Anchor Feature Selection for Single-Shot Object Detection基于动态锚特征选择单镜头目标检测 
Incremental Learning Using Conditional Adversarial Networks基于条件对抗网络增量学习 
Bilateral Adversarial Training: Towards Fast Training of More Robust Models Against Adversarial Attacks双边对抗性训练:快速训练更强大的抗对抗性攻击模型 
View Confusion Feature Learning for Person Re-Identification基于视图混淆特征学习人再识别 
Auto-FPN: Automatic Network Architecture Adaptation for Object Detection Beyond ClassificationAuto-FPN:自动网络体系结构自适应实现超越分类的目标检测 
PARN: Position-Aware Relation Networks for Few-Shot LearningPARN:基于位置感知关系网络少镜头学习 
Multi-Adversarial Faster-RCNN for Unrestricted Object Detection基于多对抗Faster-RCNN无限制目标检测 
Object Guided External Memory Network for Video Object Detection基于目标引导外存网络视频目标检测 
An Empirical Study of Spatial Attention Mechanisms in Deep Networks深度网络中空间注意机制实证研究 
Attribute Attention for Semantic Disambiguation in Zero-Shot Learning零镜头学习中基于属性注意语义消歧 
CIIDefence: Defeating Adversarial Attacks by Fusing Class-Specific Image Inpainting and Image DenoisingCIIDefence:通过融合特定类别的图像修复图像去噪战胜对抗性攻击 
ThunderNet: Towards Real-Time Generic Object Detection on Mobile DevicesThunderNet:面向移动设备的实时通用目标检测 
Dual Student: Breaking the Limits of the Teacher in Semi-Supervised Learning双重学生打破教师半监督学习中的局限 
MVP Matching: A Maximum-Value Perfect Matching for Mining Hard Samples, With Application to Person Re-IdentificationMVP匹配:挖掘难样本极大值完全匹配方法及其在人再识别中的应用 
Adaptive Context Network for Scene Parsing用于场景分析自适应上下文网络 
Constructing Self-Motivated Pyramid Curriculums for Cross-Domain Semantic Segmentation: A Non-Adversarial Approach基于自我激励金字塔课程跨域语义分割:一种非对抗性方法课程学习:基于局部分布
自我激励:基于潜变量
本文将两种方式结合起来,并结合金字塔技术,实现域自适应的语义分割
SparseMask: Differentiable Connectivity Learning for Dense Image PredictionSparseMask:用于稠密图像预测可微连通学习 
Significance-Aware Information Bottleneck for Domain Adaptive Semantic Segmentation基于重要性感知信息Bottleneck域自适应语义分割基于GAN的域自适应语义分割的改进,对潜变量进行重要性感知的限制(如图2,3)
Relational Attention Network for Crowd Counting基于关系注意力网络人群计数 
ACFNet: Attentional Class Feature Network for Semantic SegmentationACFNet:基于注意力类特征网络语义分割一种利用类别特征进行语义分割refine的方法,如图2,3。
在粗粒度的语义分割基础上,提取不同类别的特征,进一步由不同类别的特征,对骨干网提出的特征进行Attention,并在此基础上refine
Frame-to-Frame Aggregation of Active Regions in Web Videos for Weakly Supervised Semantic Segmentation基于web视频活动区域帧间聚合弱监督语义分割 
Boundary-Aware Feature Propagation for Scene Segmentation基于边界感知特征传播场景分割 
Self-Ensembling With GAN-Based Data Augmentation for Domain Adaptation in Semantic Segmentation基于GAN的数据增强自组织域自适应语义分割中的应用 
Explaining the Ambiguity of Object Detection and 6D Pose From Visual Data视觉数据解释目标检测6d姿态的模糊性 
Accurate Monocular 3D Object Detection via Color-Embedded 3D Reconstruction for Autonomous Driving基于彩色嵌入三维重建单目三维物体精确检测自动驾驶的应用 
MonoLoco: Monocular 3D Pedestrian Localization and Uncertainty Estimation单目三维行人定位不确定性估计 
Unsupervised High-Resolution Depth Learning From Videos With Dual Networks基于双网络视频无监督高分辨率深度学习 
Bayesian Graph Convolution LSTM for Skeleton Based Action Recognition贝叶斯图卷积LSTM实现基于骨架的动作识别 
DeCaFA: Deep Convolutional Cascade for Face Alignment in the WildDeCaFa:基于深度卷积级联野外人脸定位 
Probabilistic Face Embeddings概率人脸嵌入 
Gaze360: Physically Unconstrained Gaze Estimation in the WildGaze360:野外自然无约束凝视估计 
Unsupervised Person Re-Identification by Camera-Aware Similarity Consistency Learning基于摄像机感知相似一致性学习无监督人再识别 
Photo-Realistic Monocular Gaze Redirection Using Generative Adversarial Networks基于GAN单目注视重定向 
Dynamic Kernel Distillation for Efficient Pose Estimation in Videos动态核蒸馏视频位姿估计中的应用 
Single-Stage Multi-Person Pose Machines单级多人位姿机 
SO-HandNet: Self-Organizing Network for 3D Hand Pose Estimation With Semi-Supervised LearningSo-HandNet:基于自组织网络半监督三维手姿态估计 
Adaptive Wing Loss for Robust Face Alignment via Heatmap Regression利用热图回归实现基于Wing损失鲁棒人脸对齐 
Single-Network Whole-Body Pose Estimation单网络全身姿态估计 
Face Alignment With Kernel Density Deep Neural Network基于核密度深度神经网络人脸对齐 
Spatiotemporal Feature Residual Propagation for Action Prediction基于时空特征残差传播动作预测 
Identity From Here, Pose From There: Self-Supervised Disentanglement and Generation of Objects Using Unlabeled Videos从这里来的身份,从那里来的姿势:使用无标签视频自监督分离对象生成 
Relation Distillation Networks for Video Object Detection基于关系蒸馏网络视频对象检测 
Video Compression With Rate-Distortion Autoencoders基于率失真自编码器视频压缩 
Non-Local ConvLSTM for Video Compression Artifact Reduction基于非局部ConvLSTM视频压缩伪影减少 
Self-Supervised Moving Vehicle Tracking With Stereo Sound基于立体声自监督运动车辆跟踪 
Self-Supervised Learning With Geometric Constraints in Monocular Video: Connecting Flow, Depth, and Camera单目视频中带几何约束的自监督学习连接流、深度和摄像机 
Learning Temporal Action Proposals With Fewer Labels较少的标签学习时域行动建议 
TSM: Temporal Shift Module for Efficient Video UnderstandingTSM:基于时域转换模块高效视频理解 
Graph Convolutional Networks for Temporal Action Localization基于图卷积网络时域动作定位 
Fast Object Detection in Compressed Video压缩视频中快速目标检测 
Predicting 3D Human Dynamics From Video视频的三维人体动力学预测 
Imitation Learning for Human Pose Prediction基于模拟学习人体姿态预测 
Human Motion Prediction via Spatio-Temporal Inpainting基于时空修复人体运动预测 
Structured Prediction Helps 3D Human Motion Modelling结构化预测有助于三维人体运动建模 
Learning Shape Templates With Structured Implicit Functions基于结构化隐函数形状模板学习 
CompenNet++: End-to-End Full Projector CompensationCompenNet++:端到端的完整投影仪补偿 
Deep Parametric Indoor Lighting Estimation深度参数化室内照明估算 
FSGAN: Subject Agnostic Face Swapping and ReenactmentFSGAN:主体不可知的人脸交换重生成 
Deep Single-Image Portrait Relighting深度单像人像Relighting 
PU-GAN: A Point Cloud Upsampling Adversarial NetworkPU-GAN:一种点云上采样对抗网络 
Neural 3D Morphable Models: Spiral Convolutional Networks for 3D Shape Representation Learning and Generation神经三维变形模型:螺旋卷积网络三维形状表示学习与生成中的应用 
Joint Learning of Saliency Detection and Weakly Supervised Semantic Segmentation显著性检测弱监督语义分割联合学习弱监督语义分割:输入两类训练集(像素级显著性训练集和类别级分类训练集),训练后的像素级语义分割
Towards High-Resolution Salient Object Detection高分辨率显著目标检测 
Event-Based Motion Segmentation by Motion Compensation利用运动补偿实现基于事件的运动分割 
Depth-Induced Multi-Scale Recurrent Attention Network for Saliency Detection基于深度诱导多尺度递归注意力网络显著性检测 
Stacked Cross Refinement Network for Edge-Aware Salient Object Detection基于叠层交叉求精网络边缘感知显著目标检测 
Motion Guided Attention for Video Salient Object Detection基于运动引导注意力视频显著目标检测 
Semi-Supervised Video Salient Object Detection Using Pseudo-Labels基于伪标签半监督视频显著目标检测 
Joint Learning of Semantic Alignment and Object Landmark Detection语义对齐目标标志检测联合学习 
RainFlow: Optical Flow Under Rain Streaks and Rain Veiling Effect雨流:雨带雨幕效应下的光流 
GridDehazeNet: Attention-Based Multi-Scale Network for Image DehazingGridDehazeNet:基于注意力多尺度图像去雾网络 
Learning to See Moving Objects in the Dark学会在黑暗中看到移动的物体 
SegSort: Segmentation by Discriminative Sorting of SegmentsSegSort:通过判别性分段排序分割 
What Synthesis Is Missing: Depth Adaptation Integrated With Weak Supervision for Indoor Scene Parsing合成缺少什么:深度自适应弱监督相结合室内场景分析 
AdaptIS: Adaptive Instance Selection NetworkAdaptIS:自适应实例选择网络 
DADA: Depth-Aware Domain Adaptation in Semantic SegmentationDADA:基于深度感知域自适应语义分割 
Guided Curriculum Model Adaptation and Uncertainty-Aware Evaluation for Semantic Nighttime Image Segmentation基于引导课程模型自适应不确定性感知评价夜间图像语义分割课程学习、自适应、夜间图像的语义分割
SceneGraphNet: Neural Message Passing for 3D Indoor Scene AugmentationSceneGraphNet:基于神经信息传递三维室内场景增强 
SkyScapes Fine-Grained Semantic Understanding of Aerial Scenes空中场景的精细语义理解 
Transferable Representation Learning in Vision-and-Language Navigation基于可迁移表示学习视觉与语言导航 
Towards Unsupervised Image Captioning With Shared Multimodal Embeddings基于共享多模式嵌入无监督图像标注 
ViCo: Word Embeddings From Visual Co-Occurrences视觉共现实现词嵌入 
Seq-SG2SL: Inferring Semantic Layout From Scene Graph Through Sequence to Sequence LearningSeq-SG2SL:通过序列到序列学习从场景图推断语义布局 
U-CAM: Visual Explanation Using Uncertainty Based Class Activation MapsU-CAM:基于不确定性的类激活图实现可视化解释 
See-Through-Text Grouping for Referring Image Segmentation基于透明文本分组参考图像分割 
VideoBERT: A Joint Model for Video and Language Representation LearningVideoBERT:一种视频和语言表示学习联合模型 
Language Features Matter: Effective Language Representations for Vision-Language Tasks语言特征的重要性:视觉-语言任务有效语言表示 
Semantic Stereo Matching With Pyramid Cost Volumes基于金字塔CostVolume语义立体匹配1. 采用语义分割提升立体匹配
2. 采用不同尺度的CostVolume
Spatial Correspondence With Generative Adversarial Network: Learning Depth From Monocular Videos基于GAN的空间对应单目视频的学习深度 
Learning Relationships for Multi-View 3D Object Recognition基于关系学习多视图三维目标识别 
View N-Gram Network for 3D Object Retrieval基于视图N-Gram网络三维对象检索 
Expert Sample Consensus Applied to Camera Re-Localization专家样本一致性相机再定位中的应用 
Semantic Part Detection via Matching: Learning to Generalize to Novel Viewpoints From Limited Training Data基于匹配语义部分检测学习从有限的训练数据推广到新的视点 
Dynamic Points Agglomeration for Hierarchical Point Sets Learning基于动态点聚集层次点集学习 
Attributing Fake Images to GANs: Learning and Analyzing GAN Fingerprints将假图像归因于GANs:GAN指纹学习分析 
Dual Adversarial Inference for Text-to-Image Synthesis基于双对抗推理文本到图像合成 
View-LSTM: Novel-View Video Synthesis Through View Decomposition视图LSTM:一种新的基于视图分解新视图视频合成方法 
HoloGAN: Unsupervised Learning of 3D Representations From Natural ImagesHoloGAN:自然图像三维表示无监督学习 
Unpaired Image-to-Speech Synthesis With Multimodal Information Bottleneck基于多模态信息Bottleneck非配对图像-语音合成 
Improved Conditional VRNNs for Video Prediction基于条件VRNNs视频预测改进 
Visualizing the Invisible: Occluded Vehicle Segmentation and Recovery可视化看不见:遮挡车辆分割与恢复 
Learning Single Camera Depth Estimation Using Dual-Pixels利用双像素学习单摄像机深度估计 
Domain-Adaptive Single-View 3D Reconstruction域自适应单视图三维重建 
Transformable Bottleneck Networks可转换Bottleneck网络 
RIO: 3D Object Instance Re-Localization in Changing Indoor EnvironmentsRIO:在变化的室内环境中3D对象实例的重新定位 
Pix2Pose: Pixel-Wise Coordinate Regression of Objects for 6D Pose EstimationPix2Pose:基于逐像素坐标回归6D姿态估计 
CDPN: Coordinates-Based Disentangled Pose Network for Real-Time RGB-Based 6-DoF Object Pose EstimationCDPN:基于坐标的解纠缠位姿网络实现实时基于RGB的六自由度目标位姿估计 
C3DPO: Canonical 3D Pose Networks for Non-Rigid Structure From MotionC3DPO:基于标准三维位姿网络非刚性Structure From Motion 
Learning to Reconstruct 3D Manhattan Wireframes From a Single Image学习从单个图像重建曼哈顿三维线框 
Soft Rasterizer: A Differentiable Renderer for Image-Based 3D Reasoning软光栅化器:一种可微渲染器实现基于图像的三维推理 
Learnable Triangulation of Human Pose人体姿势三角剖分学习 
xR-EgoPose: Egocentric 3D Human Pose From an HMD CameraxR-EgoPose:HMD相机以自我为中心的3D人体姿势 
DeepHuman: 3D Human Reconstruction From a Single ImageDeepHuman:从单个图像重建三维人体 
A Neural Network for Detailed Human Depth Estimation From a Single Image单幅图像的人体深度精细估计神经网络 
DenseRaC: Joint 3D Pose and Shape Estimation by Dense Render-and-CompareDenseRaC:基于稠密渲染和比较联合三维姿态形状估计 
Not All Parts Are Created Equal: 3D Pose Estimation by Modeling Bi-Directional Dependencies of Body Parts并非所有的部分都是平等地创建:通过建立身体部分双向依赖关系估计三维姿势 
Extreme View Synthesis极限视图合成 
View Independent Generative Adversarial Network for Novel View Synthesis视图无关GAN新视图合成中的应用 
Cascaded Context Pyramid for Full-Resolution 3D Semantic Scene Completion基于级联上下文金字塔全分辨率三维语义场景补全 
View-Consistent 4D Light Field Superpixel Segmentation视图一致的4D光场超像素分割 
GLoSH: Global-Local Spherical Harmonics for Intrinsic Image DecompositionGLoSH:用于内在图像分解全局-局部球谐函数 
Surface Normals and Shape From Water水面法向量形状 
Restoration of Non-Rigidly Distorted Underwater Images Using a Combination of Compressive Sensing and Local Polynomial Image Representations基于压缩传感局部多项式图像表示组合非刚性畸变水下图像复原 
Learning Perspective Undistortion of Portraits学习肖像画去失真视角 
Towards Photorealistic Reconstruction of Highly Multiplexed Lensless Images高复用无透镜图像真实感重建 
Unconstrained Motion Deblurring for Dual-Lens Cameras双镜头相机的无约束运动去模糊 
Stochastic Exposure Coding for Handling Multi-ToF-Camera Interference处理多TOF相机干扰随机曝光编码 
Convolutional Approximations to the General Non-Line-of-Sight Imaging Operator一般非视线成像算子卷积逼近 
Agile Depth Sensing Using Triangulation Light Curtains基于三角光幕快速深度传感 
Asynchronous Single-Photon 3D Imaging异步单光子三维成像 
Cross-Dataset Person Re-Identification via Unsupervised Pose Disentanglement and Adaptation基于无监督姿势分离自适应跨数据集人再识别 
A Learned Representation for Scalable Vector Graphics基于表示学习可伸缩矢量图形 
ELF: Embedded Localisation of Features in Pre-Trained CNNELF:在预先训练的CNN嵌入特征定位 
Joint Group Feature Selection and Discriminative Filter Learning for Robust Visual Object Tracking基于联合组特征选择判别滤波器学习鲁棒视觉目标跟踪 
Sampling Wisely: Deep Image Embedding by Top-K Precision Optimization明智采样:基于Top-K精度优化深度图像嵌入 
On the Global Optima of Kernelized Adversarial Representation Learning核化对抗表征学习全局优化 
Addressing Model Vulnerability to Distributional Shifts Over Image Transformation Sets解决图像转换集上分布移位的模型脆弱性 
Attract or Distract: Exploit the Margin of Open Set吸引或分散注意力:探索开放集边缘 
MIC: Mining Interclass Characteristics for Improved Metric LearningMIC:挖掘类间特征改进度量学习 
Self-Supervised Representation Learning via Neighborhood-Relational Encoding基于邻域关系编码自监督表示学习 
AWSD: Adaptive Weighted Spatiotemporal Distillation for Video Representation自适应加权时空蒸馏视频表示中的应用 
Bilinear Attention Networks for Person Retrieval用于人检索双线性注意网络 
Discriminative Feature Learning With Consistent Attention Regularization for Person Re-Identification基于一致注意正则化判别特征学习用于人再识别 
Semi-Supervised Domain Adaptation via Minimax Entropy基于极大极小熵半监督域自适应 
Boosting Few-Shot Visual Learning With Self-Supervision自我监督促进少镜头视觉学习 
FDA: Feature Disruptive AttackFDA:功能破坏性攻击 
A Novel Unsupervised Camera-Aware Domain Adaptation Framework for Person Re-Identification一种新的无监督摄像机感知域自适应框架实现人再识别 
Recover and Identify: A Generative Dual Model for Cross-Resolution Person Re-Identification恢复与识别:一种生成性双重模型实现交叉分辨人再识别 
Cross-View Policy Learning for Street Navigation用于街道导航交叉视野策略学习 
Learning Across Tasks and Domains跨任务跨领域学习 
EMPNet: Neural Localisation and Mapping Using Embedded Memory PointsEMPNet:基于嵌入式存储点神经定位映射 
AVT: Unsupervised Learning of Transformation Equivariant Representations by Autoencoding Variational TransformationsAVT:自编码变分变换实现变换等变表示的无监督学习 
Composite Shape Modeling via Latent Space Factorization基于潜在空间分解复合形状建模 
Deep Comprehensive Correlation Mining for Image Clustering基于深度综合相关挖掘图像聚类 
Unsupervised Multi-Task Feature Learning on Point Clouds点云上无监督多任务特征学习 
Reciprocal Multi-Layer Subspace Learning for Multi-View Clustering基于互反多层子空间学习多视图聚类 
Geometric Disentanglement for Generative Latent Shape Models基于几何解缠生成性潜在形状模型 
GAN-Tree: An Incrementally Learned Hierarchical Generative Framework for Multi-Modal Data DistributionsGAN-Tree:一种多模态数据分布增量学习分层生成框架 
GODS: Generalized One-Class Discriminative Subspaces for Anomaly DetectionGODs:广义一类判别子空间用于异常检测 
Neighborhood Preserving Hashing for Scalable Video Retrieval可分级视频检索中的邻域保持哈希算法 
Self-Training With Progressive Augmentation for Unsupervised Cross-Domain Person Re-Identification无监督跨域人再识别渐进增强自训练 
SCRDet: Towards More Robust Detection for Small, Cluttered and Rotated ObjectsSCRDet:对小的、杂乱的和旋转物体进行更稳健的检测 
Cross-X Learning for Fine-Grained Visual Categorization基于Cross-X学习细粒度视觉分类 
Maximum-Margin Hamming Hashing最大边缘汉明散列 
Conservative Wasserstein Training for Pose Estimation基于保守Wasserstein训练姿势估计 
Learning to Rank Proposals for Object Detection基于排序建议学习目标检测 
Vehicle Re-Identification With Viewpoint-Aware Metric Learning基于视点感知度量学习车辆再识别 
WSOD2: Learning Bottom-Up and Top-Down Objectness Distillation for Weakly-Supervised Object DetectionWSPD2:基于自下而上和自上而下对象蒸馏学习弱监督对象检测 
Localization of Deep Inpainting Using High-Pass Fully Convolutional Network基于高通全卷积网络的深度修补定位 
Clustered Object Detection in Aerial Images航空图像中簇状目标检测 
Unsupervised Graph Association for Person Re-Identification基于无监督图关联人再识别 
Learning a Mixture of Granularity-Specific Experts for Fine-Grained Categorization基于粒度特定专家混合学习细粒度分类 
advPattern: Physical-World Attacks on Deep Person Re-Identification via Adversarially Transformable PatternsadvPattern:通过对抗转换模式实现对人再识别进行物理世界攻击 
ABD-Net: Attentive but Diverse Person Re-IdentificationABD-Net专注但多元的人再识别 
From Open Set to Closed Set: Counting Objects by Spatial Divide-and-Conquer从开集到闭集:基于空间分治对象计数 
Towards Precise End-to-End Weakly Supervised Object Detection Network精确的端到端弱监督目标检测网络 
Learn to Scale: Generating Multipolar Normalized Density Maps for Crowd Counting学习缩放:生成用于人群计数多极归一化密度图 
Ground-to-Aerial Image Geo-Localization With a Hard Exemplar Reweighting Triplet Loss具有难样本重加权三元损失地-空图像地理定位 
Learning to Discover Novel Visual Categories via Deep Transfer Clustering通过深度转移聚类学习发现新的视觉类别 
AM-LFS: AutoML for Loss Function SearchAM-LFS:用于损失函数搜索AutoML 
Few-Shot Object Detection via Feature Reweighting基于特征重加权少镜头目标检测 
Objects365: A Large-Scale, High-Quality Dataset for Object DetectionObjects365:用于目标检测大规模高质量数据集 
Efficient and Accurate Arbitrary-Shaped Text Detection With Pixel Aggregation Network基于像素聚集网络任意形状文本检测 
Foreground-Aware Pyramid Reconstruction for Alignment-Free Occluded Person Re-Identification基于前景感知金字塔重建无对齐遮挡人再识别 
Collect and Select: Semantic Alignment Metric Learning for Few-Shot Learning收集和选择:用于少镜头学习语义对齐度量学习 
Bayesian Adaptive Superpixel Segmentation贝叶斯自适应超像素分割 
CapsuleVOS: Semi-Supervised Video Object Segmentation Using Capsule RoutingCapsuleVOS:基于胶囊路由半监督视频对象分割 
BAE-NET: Branched Autoencoder for Shape Co-SegmentationBAE-NET:基于分支自动编码器形状共分割 
VV-Net: Voxel VAE Net With Group Convolutions for Point Cloud SegmentationVV网:基于组卷积的体素VAE网用于点云分割 
Miss Detection vs. False Alarm: Adversarial Learning for Small Object Segmentation in Infrared Images漏检与虚警:红外图像小目标分割对抗学习 
Group-Wise Deep Object Co-Segmentation With Co-Attention Recurrent Neural Network基于共注意力递归神经网络组深度目标共分割 
Human Attention in Image Captioning: Dataset and Analysis图像标注中的人注意数据集分析 
Variational Uncalibrated Photometric Stereo Under General Lighting一般光照下变分非定标光度立体 
SPLINE-Net: Sparse Photometric Stereo Through Lighting Interpolation and Normal Estimation NetworksSPLINE网:通过光插值法向估计网络稀疏光度立体 
Hyperspectral Image Reconstruction Using Deep External and Internal Learning基于内、外深度学习高光谱图像重建 
Gravity as a Reference for Estimating a Person's Height From Video参考重力实现视频中估计身高 
Shadow Removal via Shadow Image Decomposition基于阴影图像分解阴影去除 
OperatorNet: Recovering 3D Shapes From Difference OperatorsOperatorNet:从差分运算符恢复三维形状 
Neural Inverse Rendering of an Indoor Scene From a Single Image单幅图像的室内场景神经逆绘制 
ForkNet: Multi-Branch Volumetric Semantic Completion From a Single Depth ImageForkNet单深度图像的多分支体积语义补全 
Moving Indoor: Unsupervised Video Depth Learning in Challenging Environments室内移动:挑战环境下的无监督视频深度学习 
GraphX-Convolution for Point Cloud Deformation in 2D-to-3D Conversion基于GraphX卷积二维到三维转换中点云变形 
FrameNet: Learning Local Canonical Frames of 3D Surfaces From a Single RGB ImageFrameNet:从单个RGB图像学习三维曲面局部规范框架 
Holistic++ Scene Understanding: Single-View 3D Holistic Scene Parsing and Human Pose Estimation With Human-Object Interaction and Physical CommonsenseHolistic++场景理解:单视图三维整体场景解析基于人-物交互和物理常识的人体姿态估计 
MMAct: A Large-Scale Dataset for Cross Modal Human Action UnderstandingMMAct:用于跨模态人类行为理解大规模数据集 
HACS: Human Action Clips and Segments Dataset for Recognition and Temporal LocalizationHACS:用于识别时间定位人类动作片段分割数据集 
3C-Net: Category Count and Center Loss for Weakly-Supervised Action Localization3C-Net:基于类别计数中心损失弱监督动作定位 
Grounded Human-Object Interaction Hotspots From Video视频中固定的人-机交互热点 
Hallucinating IDT Descriptors and I3D Optical Flow Features for Action Recognition With CNNs利用幻觉IDT描述子I3D光流特征实现基于CNNs的动作识别 
Learning to Paint With Model-Based Deep Reinforcement Learning基于模型的深度强化学习绘画中的应用 
Neural Re-Simulation for Generating Bounces in Single Images基于神经网络再模拟的单幅图像反弹生成 
Deep Appearance Maps深度外观图 
GarNet: A Two-Stream Network for Fast and Accurate 3D Cloth DrapingGarNet:一种快速准确的三维布料覆盖双流网络 
Joint Embedding of 3D Scan and CAD Objects三维扫描CAD对象联合嵌入 
CompoNet: Learning to Generate the Unseen by Part Synthesis and CompositionCompoNet:通过部分合成组合学习生成看不见的部分 
DDSL: Deep Differentiable Simplex Layer for Learning Geometric SignalsDDSL:基于深度可微单纯形层几何信号学习 
EGNet: Edge Guidance Network for Salient Object DetectionEGNet:用于显著目标检测边缘引导网络 
SID4VAM: A Benchmark Dataset With Synthetic Images for Visual Attention ModelingSID4VAM:用于视觉注意建模合成图像基准数据集 
Two-Stream Action Recognition-Oriented Video Super-Resolution面向双流动作识别视频超分辨率 
Where Is My Mirror?我的镜子在哪里? 
Disentangled Image Matting分离图像抠图 
Guided Super-Resolution As Pixel-to-Pixel Transformation通过像素到像素转换引导超分辨率 
Deep Learning for Light Field Saliency Detection光场显著性检测深度学习 
Optimizing the F-Measure for Threshold-Free Salient Object Detection基于F-测度优化无阈值显著目标检测 
Image Inpainting With Learnable Bidirectional Attention Maps基于可学习双向注意图图像修复 
Joint Demosaicking and Denoising by Fine-Tuning of Bursts of Raw Images通过对原始图像序列的微调实现联合去马赛克去噪 
DeblurGAN-v2: Deblurring (Orders-of-Magnitude) Faster and BetterBeblurGAN-v2去模糊(数量级)更快更好 
Reflective Decoding Network for Image Captioning基于反射解码网络图像标注 
Joint Optimization for Cooperative Image Captioning协同图像标注联合优化 
Watch, Listen and Tell: Multi-Modal Weakly Supervised Dense Event Captioning看、听、说:多模弱监督密集事件标注 
Joint Syntax Representation Learning and Visual Cue Translation for Video Captioning基于联合句法表示学习视觉线索翻译视频标注 
Entangled Transformer for Image Captioning基于纠缠变换图像标注 
Shapeglot: Learning Language for Shape DifferentiationShapeglot:基于语言学习形态分化 
nocaps: novel object captioning at scalenocaps:尺度上的新对象标注 
Fully Convolutional Geometric Features完全卷积几何特征 
Learning Local RGB-to-CAD Correspondences for Object Pose Estimation基于局部RGB-CAD的对应关系学习目标姿态估计 
Depth From Videos in the Wild: Unsupervised Monocular Depth Learning From Unknown Cameras野外视频的深度:未知摄像机的无监督单目深度学习 
OmniMVS: End-to-End Learning for Omnidirectional Stereo MatchingOmniMVS:全方位立体匹配端到端学习多视角的立体匹配
On the Over-Smoothing Problem of CNN Based Disparity Estimation基于CNN的视差估计过平滑问题 
Disentangling Propagation and Generation for Video Prediction视频预测分离传播与生成 
Guided Image-to-Image Translation With Bi-Directional Feature Transformation基于双向特征变换图像-图像的转换 
Towards Multi-Pose Guided Virtual Try-On Network面向多姿态引导虚拟Try-On网络 
Photorealistic Style Transfer via Wavelet Transforms基于小波变换真实感风格转换 
Personalized Fashion Design个性化服装设计 
Tag2Pix: Line Art Colorization Using Text Tag With SECat and Changing LossTag2Pix:使用带有SECat和Changing损失的文本标记进行线条艺术着色 
Free-Form Video Inpainting With 3D Gated Convolution and Temporal PatchGAN基于三维门控卷积时域PatchGAN自由形式视频修补 
TextDragon: An End-to-End Framework for Arbitrary Shaped Text SpottingTextDragon:用于任意形状文本定位端到端框架 
Chinese Street View Text: Large-Scale Chinese Text Reading With Partially Supervised Learning中文街景文本:基于部分监督学习大规模中文文本阅读 
Deep Floor Plan Recognition Using a Multi-Task Network With Room-Boundary-Guided Attention房间边界引导注意多任务网络实现深度楼层平面识别 
GA-DAN: Geometry-Aware Domain Adaptation Network for Scene Text Detection and RecognitionGA-DAN:用于场景文本检测和识别几何感知域自适应网络 
Large-Scale Tag-Based Font Retrieval With Generative Feature Learning基于生成特征学习大规模标签字体检索 
Convolutional Character Networks卷积字符网络 
Geometry Normalization Networks for Accurate Scene Text Detection用于精确场景文本检测几何规范化网络 
Symmetry-Constrained Rectification Network for Scene Text Recognition对称约束校正网络场景文本识别中的应用 
YOLACT: Real-Time Instance SegmentationYOLACT:实时实例分割见图2,先分割出对象BB,再进行像素级实例分割
Expectation-Maximization Attention Networks for Semantic Segmentation基于期望最大化注意力网络语义分割如图2,将EM算法的思想和迭代过程,嵌入到深度网络中,目的是替代自监督Attention过程(无需访问所有数据,较Non-Local更为灵活,且可以提升速度)
Multi-Class Part Parsing With Joint Boundary-Semantic Awareness基于联合边界语义感知多类部分解析 
Explaining Neural Networks Semantically and Quantitatively神经网络语义和定量地解释 
PANet: Few-Shot Image Semantic Segmentation With Prototype AlignmentPANet:基于原型对齐少镜头图像语义分割 
ShapeMask: Learning to Segment Novel Objects by Refining Shape PriorsShapeMask:通过精化形状先验学习分割新对象 
Sequence Level Semantics Aggregation for Video Object Detection基于序列级语义聚合视频对象检测 
Video Object Segmentation Using Space-Time Memory Networks基于时空存储网络视频对象分割 
Zero-Shot Video Object Segmentation via Attentive Graph Neural Networks基于注意力图神经网络零镜头视频对象分割 
MeteorNet: Deep Learning on Dynamic 3D Point Cloud SequencesMeteorNet:动态三维点云序列深度学习 
3D Instance Segmentation via Multi-Task Metric Learning基于多任务度量学习三维实例分割 
DeepGCNs: Can GCNs Go As Deep As CNNs?DeepGCN:GCN能像CNN一样深吗 
Deep Hough Voting for 3D Object Detection in Point Clouds点云中基于深度Hough投票三维目标检测 
M3D-RPN: Monocular 3D Region Proposal Network for Object DetectionM3D-RPN:用于目标检测单目3D区域建议网络 
SemanticKITTI: A Dataset for Semantic Scene Understanding of LiDAR SequencessemanticKITTI:用于激光雷达序列语义场景理解数据集 
WoodScape: A Multi-Task, Multi-Camera Fisheye Dataset for Autonomous DrivingWoodScape:一个用于自动驾驶多任务、多摄像机鱼眼数据集 
Scalable Place Recognition Under Appearance Change for Autonomous Driving面向自主驾驶外观变化下的可扩展位置识别 
Exploring the Limitations of Behavior Cloning for Autonomous Driving探索在自主驾驶行为克隆局限性 
Habitat: A Platform for Embodied AI ResearchHabitat:体现人工智能研究的平台 
Towards Interpretable Face Recognition面向可解释的人脸识别 
Co-Mining: Deep Face Recognition With Noisy Labels联合挖掘:带噪声标签深度人脸识别 
Few-Shot Adaptive Gaze Estimation少镜头自适应注视估计 
Live Face De-Identification in Video视频中实时人脸反识别 
Face Video Deblurring Using 3D Facial Priors基于三维人脸先验视频人脸去模糊 
Semi-Supervised Monocular 3D Face Reconstruction With End-to-End Shape-Preserved Domain Transfer基于端到端形状保持域转移半监督单目三维人脸重建 
3D Face Modeling From Diverse Raw Scan Data基于多样的原始扫描数据三维人脸建模 
A Decoupled 3D Facial Shape Model by Adversarial Training一种基于对抗训练去耦三维人脸形状模型 
Photo-Realistic Facial Details Synthesis From Single Image基于单幅图像的真实感人脸细节合成 
S2GAN: Share Aging Factors Across Ages and Share Aging Trends Among IndividualsS2GAN:在各个年龄段共享老化因素,在个人间共享老化趋势 
PuppetGAN: Cross-Domain Image Manipulation by DemonstrationPuppetGAN:基于演示跨域图像操作 
Few-Shot Adversarial Learning of Realistic Neural Talking Head Models真实神经说话头部模型少镜头对抗学习 
Pose-Aware Multi-Level Feature Network for Human Object Interaction Detection基于位姿感知的多层次特征网络人机交互检测 
TRB: A Novel Triplet Representation for Understanding 2D Human BodyTRB:一种新的三元表示实现二维人体的理解 
Learning Trajectory Dependencies for Human Motion Prediction用于人体运动预测轨迹依赖学习 
Cross-Domain Adaptation for Animal Pose Estimation基于跨域自适应动物姿态估计 
NOTE-RCNN: NOise Tolerant Ensemble RCNN for Semi-Supervised Object Detection基于噪声容限集成RCNN半监督目标检测 
Unsupervised Out-of-Distribution Detection by Maximum Classifier Discrepancy基于Maximum Classifier差异无监督分布外检测 
SBSGAN: Suppression of Inter-Domain Background Shift for Person Re-IdentificationSBSGAN:基于域间背景漂移抑制人再识别 
Enriched Feature Guided Refinement Network for Object Detection基于丰富特征引导细化网络目标检测 
Deep Meta Metric Learning深度元测量学习 
Discriminative Feature Transformation for Occluded Pedestrian Detection基于判别特征变换遮挡行人检测 
Contextual Attention for Hand Detection in the Wild上下文注意野外手部检测中的应用 
Meta R-CNN: Towards General Solver for Instance-Level Low-Shot Learning元R-CNN:面向一般解算器实例级少镜头学习 
Pyramid Graph Networks With Connection Attentions for Region-Based One-Shot Semantic Segmentation基于连接注意金字塔图网络实现基于区域的单镜头语义分割 
Presence-Only Geographical Priors for Fine-Grained Image Classification用于细粒度图像分类仅存在地理先验 
POD: Practical Object Detection With Scale-Sensitive Network基于尺度敏感网络实用目标检测 
Human Uncertainty Makes Classification More Robust人类的不确定性使得分类更加可靠 
FCOS: Fully Convolutional One-Stage Object Detection全卷积单级目标检测 
Self-Critical Attention Learning for Person Re-Identification自我批判性注意力学习用于人再识别 
Temporal Knowledge Propagation for Image-to-Video Person Re-Identification基于时间知识传播图像-视频人再识别 
RepPoints: Point Set Representation for Object DetectionRepPoints:用于目标检测点集表示 
SegEQA: Video Segmentation Based Visual Attention for Embodied Question AnsweringSegEQA:一种基于视频分割视觉注意力具体问答中的应用 
No-Frills Human-Object Interaction Detection: Factorization, Layout Encodings, and Training Techniques无装饰的人机交互检测因子分解布局编码训练技术 
Cap2Det: Learning to Amplify Weak Caption Supervision for Object DetectionCap2Det:学习增强弱字幕监控以实现目标检测 
No Fear of the Dark: Image Retrieval Under Varying Illumination Conditions不怕黑暗:不同光照条件下图像检索 
Hierarchical Shot Detector分层镜头检测器 
Few-Shot Learning With Global Class Representations基于全局类表示少镜头学习 
Better to Follow, Follow to Be Better: Towards Precise Supervision of Feature Super-Resolution for Small Object Detection更好跟随,跟随更好:小目标检测特征超分辨率精确监控 
Weakly Supervised Object Detection With Segmentation Collaboration基于分割协作弱监督目标检测 
AutoFocus: Efficient Multi-Scale Inference自动聚焦:有效的多尺度推理 
Leveraging Long-Range Temporal Relationships Between Proposals for Video Object Detection基于方案之间的长范围时间关系视频对象检测 
Transferable Contrastive Network for Generalized Zero-Shot Learning基于可转移对比网络广义零镜头学习 
Fast Point R-CNN快速点R-CNN 
Mesh R-CNN网状R-CNN 
Deep Supervised Hashing With Anchor Graph基于锚图深度监督哈希算法 
Detecting 11K Classes: Large Scale Object Detection Without Fine-Grained Bounding Boxes11k类别检测:无细粒度包围盒大规模目标检测 
Re-ID Driven Localization Refinement for Person Search再识别驱动定位精化实现人搜索 
Hierarchical Encoding of Sequential Data With Compact and Sub-Linear Storage Cost基于压缩次线性存储代价序列数据分层编码 
C-MIDN: Coupled Multiple Instance Detection Network With Segmentation Guidance for Weakly Supervised Object DetectionC-MIDN:带分割指导耦合多实例检测网络实现弱监督目标检测 
Learning Feature-to-Feature Translator by Alternating Back-Propagation for Generative Zero-Shot Learning基于交替反向传播特征-特征转换学习实现零镜头学习 
Deep Constrained Dominant Sets for Person Re-Identification用于人再识别深度约束支配集 
Invariant Information Clustering for Unsupervised Image Classification and Segmentation基于不变信息聚类无监督图像分类分割 
Subspace Structure-Aware Spectral Clustering for Robust Subspace Clustering子空间结构感知谱聚类鲁棒子空间聚类中的应用 
Order-Preserving Wasserstein Discriminant Analysis保序Wasserstein判别分析 
LayoutVAE: Stochastic Scene Layout Generation From a Label SetLayoutVAE:从标签集生成随机场景布局 
Robust Variational Bayesian Point Set Registration鲁棒变分贝叶斯点集配准 
Is an Affine Constraint Needed for Affine Subspace Clustering?仿射子空间聚类需要仿射约束吗? 
Meta-Learning to Detect Rare Objects检测稀有物体元学习 
New Convex Relaxations for MRF Inference With Unknown Graphs新凸松弛实现未知图MRF推理 
Cluster Alignment With a Teacher for Unsupervised Domain Adaptation基于教师的聚类对齐实现无监督域自适应 
Analyzing the Variety Loss in the Context of Probabilistic Trajectory Prediction概率轨迹预测上下文中的变化损失分析 
Deep Mesh Reconstruction From Single RGB Images via Topology Modification Networks基于拓扑修正网络单一RGB图像深度网格重建 
UprightNet: Geometry-Aware Camera Orientation Estimation From Single ImagesUprightNet:基于单帧图像的几何感知摄像机方位估计 
Escaping Plato's Cave: 3D Shape From Adversarial Rendering逃离柏拉图的洞穴:基于对抗性渲染三维形态 
Deep End-to-End Alignment and Refinement for Time-of-Flight RGB-D Module基于深度端到端对齐与细化Time-of-Flight RGB-D模块 
GEOBIT: A Geodesic-Based Binary Descriptor Invariant to Non-Rigid Deformations for RGB-D ImagesGEOBIT:一种对RGB-D图像非刚性变形保持不变基于测地线二值描述子 
CDTB: A Color and Depth Visual Object Tracking Dataset and Benchmark彩色和深度视觉目标跟踪数据集与基准 
Learning Joint 2D-3D Representations for Depth Completion基于二维-三维联合表示学习深度补全 
Make a Face: Towards Arbitrary High Fidelity Face Manipulation做一张脸:朝向任意高保真的脸操作 
M2FPA: A Multi-Yaw Multi-Pitch High-Quality Dataset and Benchmark for Facial Pose AnalysisM2FPA:一个用于面部姿势分析多偏航多俯仰高质量数据集基准 
Fair Loss: Margin-Aware Reinforcement Learning for Deep Face Recognition公平损失:面向深度人脸识别边缘感知强化学习 
Face De-Occlusion Using 3D Morphable Model and Generative Adversarial Network基于三维变形模型GAN人脸去遮挡 
Detecting Photoshopped Faces by Scripting Photoshopphotoshop脚本检测photoshop人脸 
Ego-Pose Estimation and Forecasting As Real-Time PD Control作为实时PD控制中的自位姿估计与预测 
End-to-End Learning for Graph Decomposition图分解端到端学习 
Laplace Landmark Localization拉普拉斯地标定位 
Through-Wall Human Mesh Recovery Using Radio Signals利用无线电信号进行穿墙人体网格恢复 
Discriminatively Learned Convex Models for Set Based Face Recognition凸模型判别学习实现基于集的人脸识别 
Camera Distance-Aware Top-Down Approach for 3D Multi-Person Pose Estimation From a Single RGB Image单一RGB图像中摄像机距离感知自顶向下方法实现三维多人姿态估计 
Context-Aware Emotion Recognition Networks基于上下文感知网络情感识别 
Aggregation via Separation: Boosting Facial Landmark Detector With Semi-Supervised Style Translation基于分离的聚合:基于半监督风格平移人脸标志检测增强 
Deep Head Pose Estimation Using Synthetic Images and Partial Adversarial Domain Adaption for Continuous Label Spaces基于合成图像连续标签空间部分对抗域自适应深部头部姿态估计 
Flare in Interference-Based Hyperspectral Cameras基于干涉的高光谱相机中的耀斑 
Computational Hyperspectral Imaging Based on Dimension-Discriminative Low-Rank Tensor Recovery基于维数-判别低秩张量恢复计算高光谱成像 
Deep Optics for Monocular Depth Estimation and 3D Object Detection基于深度光学单目深度估计三维目标检测 
Physics-Based Rendering for Improving Robustness to Rain基于物理的绘制提高了对雨水的鲁棒性 
ARGAN: Attentive Recurrent Generative Adversarial Network for Shadow Detection and RemovalARGAN:用于阴影检测和消除注意力循环生成对抗网络 
Deep Tensor ADMM-Net for Snapshot Compressive Imaging用于快照压缩成像深度张量ADMM网 
Convex Relaxations for Consensus and Non-Minimal Problems in 3D Vision利用凸松弛解决三维视觉中一致性非极小问题 
Pareto Meets Huber: Efficiently Avoiding Poor Minima in Robust EstimationPareto Meets Huber:稳健估计有效避免弱极小 
K-Best Transformation SynchronizationK-最佳变换同步 
Parametric Majorization for Data-Driven Energy Minimization Methods数据驱动能量最小化方法参数优化 
A Bayesian Optimization Framework for Neural Network Compression基于贝叶斯优化框架神经网络压缩 
HiPPI: Higher-Order Projected Power Iterations for Scalable Multi-MatchingHiPPI:基于高阶投影功率迭代可伸缩多匹配 
Language-Conditioned Graph Networks for Relational Reasoning基于语言-条件图网络关系推理 
Tell, Draw, and Repeat: Generating and Modifying Images Based on Continual Linguistic Instruction讲、画、重复:基于连续语言指导图像生成修改 
Relation-Aware Graph Attention Network for Visual Question Answering基于关系感知图注意网络视觉问答 
Unpaired Image Captioning via Scene Graph Alignments基于场景图对齐未配对图像标注 
Modeling Inter and Intra-Class Relations in the Triplet Loss for Zero-Shot Learning三元损失类间和类内关系建模实现零镜头学习 
Occlusion-Shared and Feature-Separated Network for Occlusion Relationship Reasoning基于遮挡共享特征分离网络遮挡关系推理 
Compositional Video Prediction合成视频预测 
Mixture-Kernel Graph Attention Network for Situation Recognition基于混合核图注意网络态势识别 
Learning Similarity Conditions Without Explicit Supervision没有明确的监督下学习相似条件 
Joint Prediction for Kinematic Trajectories in Vehicle-Pedestrian-Mixed Scenes车-人-混合场景中运动轨迹联合预测 
Learning to Caption Images Through a Lifetime by Asking Questions通过提问来学会在一生中给图片加标注 
VrR-VG: Refocusing Visually-Relevant RelationshipsVrR-VG:重新聚焦视觉-相关关系 
TAPA-MVS: Textureless-Aware PAtchMatch Multi-View StereoTAPA-MVS:无纹理感知PatchMatch多视图立体多视图立体匹配
U4D: Unsupervised 4D Dynamic Scene UnderstandingU4D:无监督4D动态场景理解 
Hierarchical Point-Edge Interaction Network for Point Cloud Semantic Segmentation基于层次点-边交互网络点云语义分割 
Multi-Angle Point Cloud-VAE: Unsupervised Feature Learning for 3D Point Clouds From Multiple Angles by Joint Self-Reconstruction and Half-to-Half Prediction多角度点云VAE:基于联合自重构半对半预测多角度三维点云无监督特征学习 
P-MVSNet: Learning Patch-Wise Matching Confidence Aggregation for Multi-View StereoP-MVSNet:学习多视点立体视觉的逐块匹配置信聚集多视图立体匹配
SME-Net: Sparse Motion Estimation for Parametric Video Prediction Through Reinforcement Learning基于强化学习稀疏运动估计实现参数化视频预测 
ClothFlow: A Flow-Based Model for Clothed Person GenerationClothFlow:一种基于人穿衣生成模型 
LADN: Local Adversarial Disentangling Network for Facial Makeup and De-MakeupLADN:用于面部化妆和卸妆局部对抗分离网络 
Point-to-Point Video Generation点对点视频生成 
Semantics-Enhanced Adversarial Nets for Text-to-Image Synthesis基于语义增强对抗网文本-图像合成 
VTNFP: An Image-Based Virtual Try-On Network With Body and Clothing Feature PreservationVTNFP:一种身体和衣服特征保持基于图像的虚拟试穿网络 
Boundless: Generative Adversarial Networks for Image ExtensionBoundless:基于GAN图像扩展 
Image Synthesis From Reconfigurable Layout and Style基于可重构布局和风格图像合成 
Attribute Manipulation Generative Adversarial Networks for Fashion Images基于属性操作GAN时尚图像 
Few-Shot Unsupervised Image-to-Image Translation少镜头无监督图像-图像的转换 
Very Long Natural Scenery Image Prediction by Outpainting利用Outpainting实现超长自然景物图像预测 
Scaling Recurrent Models via Orthogonal Approximations in Tensor Trains张量训练中利用正交逼近实现递推模型分级 
A Deep Cybersickness Predictor Based on Brain Signal Analysis for Virtual Reality Contents虚拟现实内容中基于脑信号分析深度晕机预测 
Learning With Unsure Data for Medical Image Diagnosis医学影像诊断中的不确定性数据学习 
Recursive Cascaded Networks for Unsupervised Medical Image Registration基于递归级联网络无监督医学图像配准 
DUAL-GLOW: Conditional Flow-Based Generative Model for Modality TransferDUAL-GLOW:基于条件流生成模型实现模态转换 
Dilated Convolutional Neural Networks for Sequential Manifold-Valued Data扩张卷积神经网络用于序列流形-值数据 
Align, Attend and Locate: Chest X-Ray Diagnosis via Contrast Induced Attention Network With Limited Supervision对齐、出席和定位:有限监督下通过造影诱导注意网络进行胸部x线诊断 
Joint Acne Image Grading and Counting via Label Distribution Learning基于标签分布学习痤疮图像联合分级计数 
An Alarm System for Segmentation Algorithm Based on Shape Model基于形状模型分割算法报警系统 
HistoSegNet: Semantic Segmentation of Histological Tissue Type in Whole Slide ImagesHistoSegNet:全幻灯片图像组织类型语义分割 
Prior-Aware Neural Network for Partially-Supervised Multi-Organ Segmentation基于先验感知神经网络部分监督多器官分割 
CAMEL: A Weakly Supervised Learning Framework for Histopathology Image SegmentationCAMEL:组织病理学图像分割弱监督学习框架 
Conditional Recurrent Flow: Conditional Generation of Longitudinal Samples With Applications to Neuroimaging条件返流:纵向样本的条件生成及其在神经影像学中的应用 
Multi-Stage Pathological Image Classification Using Semantic Segmentation基于语义分割多阶段病理图像分类 
Semantic-Transferable Weakly-Supervised Endoscopic Lesions Segmentation语义可转移弱监督内镜病变分割 
Unsupervised Microvascular Image Segmentation Using an Active Contours Mimicking Neural Network基于活动轮廓模拟神经网络的无监督微血管图像分割 
GLAMpoints: Greedily Learned Accurate Match PointsGLAMpoints:贪婪地学习精确的匹配点 

 

声明:本文内容由网友自发贡献,不代表【wpsshop博客】立场,版权归原作者所有,本站不承担相应法律责任。如您发现有侵权的内容,请联系我们。转载请注明出处:https://www.wpsshop.cn/w/花生_TL007/article/detail/90880
推荐阅读
相关标签
  

闽ICP备14008679号