计算机视觉论文-2021-06-24_fairness for image generation with uncertain sensi

本专栏是计算机视觉方向论文收集积累，时间：2021年6月24日，来源：paper digest

欢迎关注原创公众号 【计算机视觉联盟】，回复 【西瓜书手推笔记】 可获取我的机器学习纯手推笔记！

直达笔记地址：机器学习手推笔记（GitHub地址）

1, TITLE: Alias-Free Generative Adversarial Networks
AUTHORS: TERO KARRAS et. al.
CATEGORY: cs.CV [cs.CV, cs.AI, cs.LG, cs.NE, stat.ML]
HIGHLIGHT: We observe that despite their hierarchical convolutional nature, the synthesis process of typical generative adversarial networks depends on absolute pixel coordinates in an unhealthy manner.

2, TITLE: Automatic Head Overcoat Thickness Measure with NASNet-Large-Decoder Net
AUTHORS: YOUSHAN ZHANG et. al.
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: To measure the thickness of the segmented HOC layer, we propose a regressive convolutional neural network (RCNN) model as well as orthogonal thickness calculation methods.

3, TITLE: Volume Rendering of Neural Implicit Surfaces
AUTHORS: Lior Yariv ; Jiatao Gu ; Yoni Kasten ; Yaron Lipman
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: The goal of this paper is to improve geometry representation and reconstruction in neural volume rendering.

4, TITLE: On Matrix Factorizations in Subspace Clustering
AUTHORS: Reeshad Arian ; Keaton Hamm
CATEGORY: cs.CV [cs.CV, 68P99, 68T10, 62H30]
HIGHLIGHT: This article explores subspace clustering algorithms using CUR decompositions, and examines the effect of various hyperparameters in these algorithms on clustering performance on two real-world benchmark datasets, the Hopkins155 motion segmentation dataset and the Yale face dataset.

5, TITLE: Generative Self-training for Cross-domain Unsupervised Tagged-to-Cine MRI Synthesis
AUTHORS: XIAOFENG LIU et. al.
CATEGORY: cs.CV [cs.CV, cs.AI, cs.LG, cs.NE]
HIGHLIGHT: In this work, we propose a novel generative self-training (GST) UDA framework with continuous value prediction and regression objective for cross-domain image synthesis.

6, TITLE: Transfer Learning of Deep Spatiotemporal Networks to Model Arbitrarily Long Videos of Seizures
AUTHORS: Fernando P�rez-Garc�a ; Catherine Scott ; Rachel Sparks ; Beate Diehl ; S�bastien Ourselin
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: We present GESTURES, a novel architecture combining convolutional neural networks (CNNs) and recurrent neural networks (RNNs) to learn deep representations of arbitrarily long videos of epileptic seizures.

7, TITLE: Adapting Off-the-Shelf Source Segmenter for Target Medical Image Segmentation
AUTHORS: Xiaofeng Liu ; Fangxu Xing ; Chao Yang ; Georges El Fakhri ; Jonghye Woo
CATEGORY: cs.CV [cs.CV, cs.AI, cs.LG]
HIGHLIGHT: To alleviate this, in this work, we target source free UDA for segmentation, and propose to adapt an ``off-the-shelf" segmentation model pre-trained in the source domain to the target domain, with an adaptive batch-wise normalization statistics adaptation framework.

8, TITLE: Vision-based Behavioral Recognition of Novelty Preference in Pigs
AUTHORS: ANIKET SHIRKE et. al.
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: We introduce a subset of such videos in the form of the 'Pig Novelty Preference Behavior' (PNPB) dataset that is fully annotated with pig actions and keypoints.

9, TITLE: The Neurally-Guided Shape Parser: A Monte Carlo Method for Hierarchical Labeling of Over-segmented 3D Shapes
AUTHORS: R. Kenny Jones ; Rana Hanocka ; Daniel Ritchie
CATEGORY: cs.CV [cs.CV, cs.AI, cs.LG]
HIGHLIGHT: In this paper, we investigate this second claim by presenting the Neurally-Guided Shape Parser (NGSP), a method that learns how to assign semantic labels to regions of an over-segmented 3D shape.

10, TITLE: Instance-based Vision Transformer for Subtyping of Papillary Renal Cell Carcinoma in Histopathological Image
AUTHORS: ZEYU GAO et. al.
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: This paper proposes a novel instance-based Vision Transformer (i-ViT) to learn robust representations of histopathological images for the pRCC subtyping task by extracting finer features from instance patches (by cropping around segmented nuclei and assigning predicted grades).

11, TITLE: Team PyKale (xy9) Submission to The EPIC-Kitchens 2021 Unsupervised Domain Adaptation Challenge for Action Recognition
AUTHORS: Xianyuan Liu ; Raivo Koot ; Shuo Zhou ; Tao Lei ; Haiping Lu
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: This report describes the technical details of our submission to the EPIC-Kitchens 2021 Unsupervised Domain Adaptation Challenge for Action Recognition.

12, TITLE: P2T: Pyramid Pooling Transformer for Scene Understanding
AUTHORS: Yu-Huan Wu ; Yun Liu ; Xin Zhan ; Ming-Ming Cheng
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: Hence, we propose to adapt pyramid pooling to MHSA for alleviating its high requirement on computational resources (problem i)).

13, TITLE: Mutual-Information Based Few-Shot Classification
AUTHORS: MALIK BOUDIAF et. al.
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: We introduce Transductive Infomation Maximization (TIM) for few-shot learning.

14, TITLE: PatentNet: A Large-Scale Incomplete Multiview, Multimodal, Multilabel Industrial Goods Image Database
AUTHORS: FANGYUAN LEI et. al.
CATEGORY: cs.CV [cs.CV, cs.AI]
HIGHLIGHT: In this paper, we introduce an industrial goods dataset, namely PatentNet, with numerous highly diverse, accurate and detailed annotations of industrial goods images, and corresponding texts.

15, TITLE: Estimating The Robustness of Classification Models By The Structure of The Learned Feature-Space
AUTHORS: Kalun Ho ; Franz-Josef Pfreundt ; Janis Keuper ; Margret Keuper
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: We introduce robustness indicators which are obtained via unsupervised clustering of latent representations inside a trained classifier and show very high correlations to the model performance on corrupted test data.

16, TITLE: 3D Human Tongue Reconstruction from Single "in-the-wild" Images
AUTHORS: Stylianos Ploumpis ; Stylianos Moschoglou ; Vasileios Triantafyllou ; Stefanos Zafeiriou
CATEGORY: cs.CV [cs.CV, cs.AI, cs.GR]
HIGHLIGHT: In this work we present the first, to the best of our knowledge, end-to-end trainable pipeline that accurately reconstructs the 3D face together with the tongue.

17, TITLE: Image-to-Image Translation of Synthetic Samples for Rare Classes
AUTHORS: Edoardo Lanzini ; Sara Beery
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: We explore the use of image-to-image translation methods to close the domain gap between synthetic and real imagery for animal species classification in data collected from camera traps: motion-activated static cameras used to monitor wildlife.

18, TITLE: Co-advise: Cross Inductive Bias Distillation
AUTHORS: SUCHENG REN et. al.
CATEGORY: cs.CV [cs.CV, cs.LG]
HIGHLIGHT: To make it into practical utility, we propose a novel distillation-based method to train vision transformers.

19, TITLE: Multi-Class Classification of Blood Cells -- End to End Computer Vision Based Diagnosis Case Study
AUTHORS: Sai Sukruth Bezugam
CATEGORY: cs.CV [cs.CV, cs.AI, q-bio.CB, stat.ML]
HIGHLIGHT: In this work we tackle the problem of white blood cell classification based on the morphological characteristics of their outer contour, color.

20, TITLE: A New Video Synopsis Based Approach Using Stereo Camera
AUTHORS: Talha Dilber ; Mehmet Serdar Guzel ; Erkan Bostanci
CATEGORY: cs.CV [cs.CV, eess.IV]
HIGHLIGHT: The model we developed has been tested and verified separately for single camera and dual camera systems.

21, TITLE: How Well Do Feature Visualizations Support Causal Understanding of CNN Activations?
AUTHORS: ROLAND S. ZIMMERMANN et. al.
CATEGORY: cs.CV [cs.CV, cs.AI, cs.HC, cs.LG]
HIGHLIGHT: Taken together, we propose an objective psychophysical task to quantify the benefit of unit-level interpretability methods for humans, and find no evidence that feature visualizations provide humans with better "causal understanding" than simple alternative visualizations.

22, TITLE: Real-time Instance Segmentation with Discriminative Orientation Maps
AUTHORS: WENTAO DU et. al.
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this paper, we propose a real-time instance segmentation framework termed OrienMask.

23, TITLE: Sentinel-1 and Sentinel-2 Spatio-Temporal Data Fusion for Clouds Removal
AUTHORS: ALESSANDRO SEBASTIANELLI et. al.
CATEGORY: cs.CV [cs.CV, cs.AI, eess.IV]
HIGHLIGHT: In this manuscript, a novel method for clouds-corrupted optical image restoration has been presented and developed, based on a joint data fusion paradigm, where three deep neural networks have been combined in order to fuse spatio-temporal features extracted from Sentinel-1 and Sentinel-2 time-series of data.

24, TITLE: Vision Permutator: A Permutable MLP-Like Architecture for Visual Recognition
AUTHORS: QIBIN HOU et. al.
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this paper, we present Vision Permutator, a conceptually simple and data efficient MLP-like architecture for visual recognition.

25, TITLE: FusionPainting: Multimodal Fusion with Adaptive Attention for 3D Object Detection
AUTHORS: SHAOQING XU et. al.
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this work, we propose a general multimodal fusion framework FusionPainting to fuse the 2D RGB image and 3D point clouds at a semantic level for boosting the 3D object detection task.

26, TITLE: Fairness in Cardiac MR Image Analysis: An Investigation of Bias Due to Data Imbalance in Deep Learning Based Segmentation
AUTHORS: ESTHER PUYOL-ANTON et. al.
CATEGORY: cs.CV [cs.CV, cs.AI]
HIGHLIGHT: In this work, we perform such an analysis for racial/gender groups, focusing on the problem of training data imbalance, using a nnU-Net model trained and evaluated on cine short axis cardiac MR data from the UK Biobank dataset, consisting of 5,903 subjects from 6 different racial groups.

27, TITLE: A Circular-Structured Representation for Visual Emotion Distribution Learning
AUTHORS: Jingyuan Yang ; Ji Lie ; Leida Li ; Xiumei Wang ; Xinbo Gao
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: Inspired by this, we propose a well-grounded circular-structured representation to utilize the prior knowledge for visual emotion distribution learning.

28, TITLE: Neural Fashion Image Captioning : Accounting for Data Diversity
AUTHORS: Gilles Hacheme ; Noureini Sayouti
CATEGORY: cs.CV [cs.CV, cs.AI, I.2.10; I.2.7; I.4.10]
HIGHLIGHT: To contribute addressing dataset diversity issues, we introduced the InFashAIv1 dataset containing almost 16.000 African fashion item images with their titles, prices and general descriptions.

29, TITLE: Transformer Meets Convolution: A Bilateral Awareness Net-work for Semantic Segmentation of Very Fine Resolution Ur-ban Scene Images
AUTHORS: LIBO WANG et. al.
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this manuscript, we pro-pose a bilateral awareness network (BANet) which contains a dependency path and a texture path to fully capture the long-range relationships and fine-grained details in VFR images.

30, TITLE: Bootstrap Representation Learning for Segmentation on Medical Volumes and Sequences
AUTHORS: Zejian Chen ; Wei Zhuo ; Tianfu Wang ; Wufeng Xue ; Dong Ni
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this work, we propose a novel straightforward method for medical volume and sequence segmentation with limited annotations.

31, TITLE: Exploiting Negative Learning for Implicit Pseudo Label Rectification in Source-Free Domain Adaptive Semantic Segmentation
AUTHORS: XIN LUO et. al.
CATEGORY: cs.CV [cs.CV, cs.AI]
HIGHLIGHT: Aiming at these pitfalls, this study develops a domain adaptive solution to semantic segmentation with pseudo label rectification (namely \textit{PR-SFDA}), which operates in two phases: 1) \textit{Confidence-regularized unsupervised learning}: Maximum squares loss applies to regularize the target model to ensure the confidence in prediction; and 2) \textit{Noise-aware pseudo label learning}: Negative learning enables tolerance to noisy pseudo labels in training, meanwhile positive learning achieves fast convergence.

32, TITLE: A Label Management Mechanism for Retinal Fundus Image Classification of Diabetic Retinopathy
AUTHORS: MENGDI GAO et. al.
CATEGORY: cs.CV [cs.CV, cs.AI]
HIGHLIGHT: In this work, we propose a novel label management mechanism (LMM) for the DNN to overcome overfitting on the noisy data.

33, TITLE: Deep Unsupervised 3D Human Body Reconstruction from A Sparse Set of Landmarks
AUTHORS: Meysam Madadi ; Hugo Bertiche ; Sergio Escalera
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this paper we propose the first deep unsupervised approach in human body reconstruction to estimate body surface from a sparse set of landmarks, so called DeepMurf.

34, TITLE: Reachability Analysis of Convolutional Neural Networks
AUTHORS: XIAODONG YANG et. al.
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: To address this challenge, we propose an approach to compute the exact reachable sets of a network given an input domain, where the reachable set is represented by the face lattice structure.

35, TITLE: LegoFormer: Transformers for Block-by-Block Multi-view 3D Reconstruction
AUTHORS: Farid Yagubbayli ; Alessio Tonioni ; Federico Tombari
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: We propose LegoFormer, a transformer-based model that unifies object reconstruction under a single framework and parametrizes the reconstructed occupancy grid by its decomposition factors.

36, TITLE: Region-Aware Network: Model Human's Top-Down Visual Perception Mechanism for Crowd Counting
AUTHORS: YUEHAI CHEN et. al.
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: Hence, in this paper, we propose a novel feedback network with Region-Aware block called RANet by modeling human's Top-Down visual perception mechanism.

37, TITLE: Gradient-Based Interpretability Methods and Binarized Neural Networks
AUTHORS: Amy Widdicombe ; Simon J. Julier
CATEGORY: cs.CV [cs.CV, cs.AI, cs.LG]
HIGHLIGHT: In this paper, we compare the performance of several widely used saliency map-based interpretabilty techniques (Gradient, SmoothGrad and GradCAM), when applied to Binarized or Full Precision Neural Networks (FPNNs).

38, TITLE: Open Images V5 Text Annotation and Yet Another Mask Text Spotter
AUTHORS: Ilya Krylov ; Sergei Nosov ; Vladislav Sovrasov
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this paper we present text annotation for Open Images V5 dataset.

39, TITLE: APNN-TC: Accelerating Arbitrary Precision Neural Networks on Ampere GPU Tensor Cores
AUTHORS: Boyuan Feng ; Yuke Wang ; Tong Geng ; Ang Li ; Yufei Ding
CATEGORY: cs.DC [cs.DC, cs.AI, cs.AR, cs.CV]
HIGHLIGHT: To break such restrictions, we introduce the first Arbitrary Precision Neural Network framework (APNN-TC) to fully exploit quantization benefits on Ampere GPU Tensor Cores.

40, TITLE: A Review of Assistive Technologies for Activities of Daily Living of Elderly
AUTHORS: Nirmalya Thakur ; Chia Y. Han
CATEGORY: cs.HC [cs.HC, cs.AI, cs.CV, cs.LG]
HIGHLIGHT: To address these needs, this work consists of making three major contributions in this field.

41, TITLE: Fairness for Image Generation with Uncertain Sensitive Attributes
AUTHORS: Ajil Jalal ; Sushrut Karmalkar ; Jessica Hoffmann ; Alexandros G. Dimakis ; Eric Price
CATEGORY: cs.LG [cs.LG, cs.CV, stat.ML]
HIGHLIGHT: This work tackles the issue of fairness in the context of generative procedures, such as image super-resolution, which entail different definitions from the standard classification setting.

42, TITLE: Multiband VAE: Latent Space Partitioning for Knowledge Consolidation in Continual Learning
AUTHORS: Kamil Deja ; Pawe? Wawrzy?ski ; Daniel Marczak ; Wojciech Masarczyk ; Tomasz Trzci?ski
CATEGORY: cs.LG [cs.LG, cs.CV]
HIGHLIGHT: We propose a new method for unsupervised continual knowledge consolidation in generative models that relies on the partitioning of Variational Autoencoder's latent space.

43, TITLE: Feature Alignment for Approximated Reversibility in Neural Networks
AUTHORS: Tiago de Souza Farias ; Jonas Maziero
CATEGORY: cs.LG [cs.LG, cs.CV]
HIGHLIGHT: We introduce feature alignment, a technique for obtaining approximate reversibility in artificial neural networks.

44, TITLE: Learning Multimodal VAEs Through Mutual Supervision
AUTHORS: TOM JOY et. al.
CATEGORY: cs.LG [cs.LG, cs.CV]
HIGHLIGHT: Here we introduce a novel alternative, the MEME, that avoids such explicit combinations by repurposing semi-supervised VAEs to combine information between modalities implicitly through mutual supervision.

45, TITLE: Behavior Mimics Distribution: Combining Individual and Group Behaviors for Federated Learning
AUTHORS: Hua Huang ; Fanhua Shang ; Yuanyuan Liu ; Hongying Liu
CATEGORY: cs.LG [cs.LG, cs.CV, cs.DC]
HIGHLIGHT: This paper proposes a novel Federated Learning algorithm (called IGFL), which leverages both Individual and Group behaviors to mimic distribution, thereby improving the ability to deal with heterogeneity.

46, TITLE: Towards Consistent Predictive Confidence Through Fitted Ensembles
AUTHORS: Navid Kardan ; Ankit Sharma ; Kenneth O. Stanley
CATEGORY: cs.LG [cs.LG, cs.AI, cs.CV]
HIGHLIGHT: This paper introduces separable concept learning framework to realistically measure the performance of classifiers in presence of OOD examples.

47, TITLE: Collaborative Visual Inertial SLAM for Multiple Smart Phones
AUTHORS: Jialing Liu ; Ruyu Liu ; Kaiqi Chen ; Jianhua Zhang ; Dongyan Guo
CATEGORY: cs.RO [cs.RO, cs.CV]
HIGHLIGHT: We propose a multi-intelligence collaborative monocular visual-inertial SLAM deployed on multiple ios mobile devices with a centralized architecture.

48, TITLE: Coarse-to-Fine Q-attention: Efficient Learning for Visual Robotic Manipulation Via Discretisation
AUTHORS: Stephen James ; Kentaro Wada ; Tristan Laidlow ; Andrew J. Davison
CATEGORY: cs.RO [cs.RO, cs.AI, cs.CV, cs.LG]
HIGHLIGHT: We therefore propose to apply this voxel prediction in a coarse-to-fine manner by gradually increasing the resolution.

49, TITLE: Multi-modal and Frequency-weighted Tensor Nuclear Norm for Hyperspectral Image Denoising
AUTHORS: Sheng Liu ; Xiaozhen Xie ; Wenfeng Kong ; Jifeng Ning
CATEGORY: eess.IV [eess.IV, cs.CV]
HIGHLIGHT: In this paper, we propose the multi-modal and frequency-weighted tensor nuclear norm (MFWTNN) and the non-convex MFWTNN for HSI denoising tasks.

50, TITLE: STRESS: Super-Resolution for Dynamic Fetal MRI Using Self-Supervised Learning
AUTHORS: Junshen Xu ; Esra Abaci Turk ; P. Ellen Grant ; Polina Golland ; Elfar Adalsteinsson
CATEGORY: eess.IV [eess.IV, cs.CV]
HIGHLIGHT: To address this problem, in this work, we propose STRESS (Spatio-Temporal Resolution Enhancement with Simulated Scans), a self-supervised super-resolution framework for dynamic fetal MRI with interleaved slice acquisitions.

51, TITLE: High-Throughput Precision Phenotyping of Left Ventricular Hypertrophy with Cardiovascular Deep Learning
AUTHORS: GRANT DUFFY et. al.
CATEGORY: eess.IV [eess.IV, cs.CV, cs.LG]
HIGHLIGHT: To overcome this challenge, we present EchoNet-LVH - a deep learning workflow that automatically quantifies ventricular hypertrophy with precision equal to human experts and predicts etiology of LVH.

52, TITLE: FoldIt: Haustral Folds Detection and Segmentation in Colonoscopy Videos
AUTHORS: Shawn Mathew ; Saad Nadeem ; Arie Kaufman
CATEGORY: eess.IV [eess.IV, cs.CV]
HIGHLIGHT: We present a novel generative adversarial network, FoldIt, for feature-consistent image translation of optical colonoscopy videos to virtual colonoscopy renderings with haustral fold overlays.

53, TITLE: Learning from Pseudo Lesion: A Self-supervised Framework for COVID-19 Diagnosis
AUTHORS: Zhongliang Li ; Zhihao Jin ; Xuechen Li ; Linlin Shen
CATEGORY: eess.IV [eess.IV, cs.CV]
HIGHLIGHT: Inspired by Ground Glass Opacity (GGO), a common finding in COIVD-19 patient's CT scans, we proposed in this paper a novel self-supervised pretraining method based on pseudo lesions generation and restoration for COVID-19 diagnosis.

54, TITLE: CxSE: Chest X-ray Slow Encoding CNN ForCOVID-19 Diagnosis
AUTHORS: Thangarajah Akilan
CATEGORY: eess.IV [eess.IV, cs.CV]
HIGHLIGHT: This work proposes a new convolutional neural network (CNN) architecture called 'slow Encoding CNN.

55, TITLE: Deformed2Self: Self-Supervised Denoising for Dynamic Medical Imaging
AUTHORS: Junshen Xu ; Elfar Adalsteinsson
CATEGORY: eess.IV [eess.IV, cs.CV]
HIGHLIGHT: In this work, we propose Deformed2Self, an end-to-end self-supervised deep learning framework for dynamic imaging denoising.