计算机视觉论文-2021-07-21

本专栏是计算机视觉方向论文收集积累，时间：2021年7月21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: DSP: Dual Soft-Paste for Unsupervised Domain Adaptive Semantic Segmentation
AUTHORS: Li Gao ; Jing Zhang ; Lefei Zhang ; Dacheng Tao
CATEGORY: cs.CV [cs.CV, cs.AI]
HIGHLIGHT: To address this issue, we propose a novel Dual Soft-Paste (DSP) method in this paper.

2, TITLE: Built-in Elastic Transformations for Improved Robustness
AUTHORS: Sadaf Gulshad ; Ivan Sosnovik ; Arnold Smeulders
CATEGORY: cs.CV [cs.CV, cs.AI, cs.LG]
HIGHLIGHT: In this paper, we start from elastic perturbations, which approximate (local) view-point changes of the object.

3, TITLE: DeepSocNav: Social Navigation By Imitating Human Behaviors
AUTHORS: Juan Pablo de Vicente ; Alvaro Soto
CATEGORY: cs.CV [cs.CV, cs.AI, cs.LG, cs.RO]
HIGHLIGHT: In this work, we propose a strategy to exploit the power of current game engines, such as Unity, to transform pre-existing bird's-eye view datasets into a first-person view, in particular, a depth view.

4, TITLE: Examining The Human Perceptibility of Black-Box Adversarial Attacks on Face Recognition
AUTHORS: Benjamin Spetter-Goldstein ; Nataniel Ruiz ; Sarah Adel Bargal
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: Through examining and measuring both the effectiveness of recent black-box attacks in the face recognition setting and their corresponding human perceptibility through survey data, we demonstrate the trade-offs in perceptibility that occur as attacks become more aggressive.

5, TITLE: ReSSL: Relational Self-Supervised Learning with Weak Augmentation
AUTHORS: MINGKAI ZHENG et. al.
CATEGORY: cs.CV [cs.CV, cs.LG]
HIGHLIGHT: In this paper, we introduced a novel SSL paradigm, which we term as relational self-supervised learning (ReSSL) framework that learns representations by modeling the relationship between different instances.

6, TITLE: Cell Detection from Imperfect Annotation By Pseudo Label Selection Using P-classification
AUTHORS: Kazuma Fujii ; Suehiro Daiki ; Nishimura Kazuya ; Bise Ryoma
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: We treat partially labeled cells as positive samples and the detected positions except for the labeled cell as unlabeled samples.

7, TITLE: Data Hiding with Deep Learning: A Survey Unifying Digital Watermarking and Steganography
AUTHORS: OLIVIA BYRNES et. al.
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: Steganography, another form of data hiding, embeds data for the purpose of secure and secret communication.

8, TITLE: Locality-aware Channel-wise Dropout for Occluded Face Recognition
AUTHORS: MINGJIE HE et. al.
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this paper, based on the argument that the occlusion essentially damages a group of neurons, we propose a novel and elegant occlusion-simulation method via dropping the activations of a group of neurons in some elaborately selected channel.

9, TITLE: SynthTIGER: Synthetic Text Image GEneratoR Towards Better Text Recognition Models
AUTHORS: Moonbin Yim ; Yoonsik Kim ; Han-Cheol Cho ; Sungrae Park
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this paper, we introduce a new synthetic text image generator, SynthTIGER, by analyzing techniques used for text image synthesis and integrating effective ones under a single algorithm.

10, TITLE: Self-Supervised Domain Adaptation for Diabetic Retinopathy Grading Using Vessel Image Reconstruction
AUTHORS: Duy M. H. Nguyen ; Truong T. N. Mai ; Ngoc T. T. Than ; Alexander Prange ; Daniel Sonntag
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: This paper investigates the problem of domain adaptation for diabetic retinopathy (DR) grading.

11, TITLE: Audio2Head: Audio-driven One-shot Talking-head Generation with Natural Head Motion
AUTHORS: Suzhen Wang ; Lincheng Li ; Yu Ding ; Changjie Fan ; Xin Yu
CATEGORY: cs.CV [cs.CV, cs.CL]
HIGHLIGHT: We propose an audio-driven talking-head method to generate photo-realistic talking-head videos from a single reference image.

12, TITLE: Discriminator-Free Generative Adversarial Attack
AUTHORS: SHAOHAO LU et. al.
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: Thegenerative-based adversarial attacks can get rid of this limitation,and some relative works propose the approaches based on GAN.However, suffering from the difficulty of the convergence of train-ing a GAN, the adversarial examples have either bad attack abilityor bad visual quality.

13, TITLE: QVHighlights: Detecting Moments and Highlights in Videos Via Natural Language Queries
AUTHORS: Jie Lei ; Tamara L. Berg ; Mohit Bansal
CATEGORY: cs.CV [cs.CV, cs.AI, cs.CL]
HIGHLIGHT: To address this issue, we present the Query-based Video Highlights (QVHighlights) dataset.

14, TITLE: Accelerating Deep Neural Networks for Efficient Scene Understanding in Automotive Cyber-physical Systems
AUTHORS: Stavros Nousias ; Erion-Vasilis Pikoulis ; Christos Mavrokefalidis ; Aris S. Lalos
CATEGORY: cs.CV [cs.CV, cs.LG]
HIGHLIGHT: Our goal in this work is to investigate best practices for appropriately applying novel weight sharing techniques, optimizing the available variables and the training procedures towards the significant acceleration of widely adopted DNNs.

15, TITLE: Understanding Gender and Racial Disparities in Image Recognition Models
AUTHORS: Rohan Mahadev ; Anindya Chakravarti
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: We investigate one such approach - which uses a multi-label softmax loss with cross-entropy as the loss function instead of a binary cross-entropy on a multi-label classification problem on the Inclusive Images dataset which is a subset of the OpenImages V6 dataset.

16, TITLE: Active 3D Shape Reconstruction from Vision and Touch
AUTHORS: EDWARD J. SMITH et. al.
CATEGORY: cs.CV [cs.CV, cs.RO]
HIGHLIGHT: In this paper, we focus on this problem and introduce a system composed of: 1) a haptic simulator leveraging high spatial resolution vision-based tactile sensors for active touching of 3D objects; 2) a mesh-based 3D shape reconstruction model that relies on tactile or visuotactile signals; and 3) a set of data-driven solutions with either tactile or visuotactile priors to guide the shape exploration.

17, TITLE: Separating Skills and Concepts for Novel Visual Question Answering
AUTHORS: Spencer Whitehead ; Hui Wu ; Heng Ji ; Rogerio Feris ; Kate Saenko
CATEGORY: cs.CV [cs.CV, cs.CL, cs.LG]
HIGHLIGHT: We present a novel method for learning to compose skills and concepts that separates these two factors implicitly within a model by learning grounded concept representations and disentangling the encoding of skills from that of concepts.

18, TITLE: Saliency for Free: Saliency Prediction As A Side-effect of Object Recognition
AUTHORS: Carola Figueroa-Flores ; David Berga ; Joost van der Weijer ; Bogdan Raducanu
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In the current paper, we demonstrate that saliency maps can be generated as a side-effect of training an object recognition deep neural network that is endowed with a saliency branch.

19, TITLE: S2Looking: A Satellite Side-Looking Dataset for Building Change Detection
AUTHORS: LI SHEN et. al.
CATEGORY: cs.CV [cs.CV, cs.AI, eess.IV]
HIGHLIGHT: In this paper, we introduce S2Looking, a building change detection dataset that contains large-scale side-looking satellite images captured at varying off-nadir angles.

20, TITLE: Learning A Sensor-invariant Embedding of Satellite Data: A Case Study for Lake Ice Monitoring
AUTHORS: Manu Tom ; Yuchang Jiang ; Emmanuel Baltsavias ; Konrad Schindler
CATEGORY: cs.CV [cs.CV, eess.IV]
HIGHLIGHT: Here, we explore the joint analysis of imagery from different sensors in the light of representation learning: we propose to learn a joint, sensor-invariant embedding (feature representation) within a deep neural network.

21, TITLE: Boosting Few-shot Classification with View-learnable Contrastive Learning
AUTHORS: Xu Luo ; Yuxuan Chen ; Liangjian Wen ; Lili Pan ; Zenglin Xu
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: To tackle this problem, we introduce the contrastive loss into few-shot classification for learning latent fine-grained structure in the embedding space.

22, TITLE: Test-Agnostic Long-Tailed Recognition By Test-Time Aggregating Diverse Experts with Self-Supervision
AUTHORS: Yifan Zhang ; Bryan Hooi ; Lanqing Hong ; Jiashi Feng
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this work, we study a more practical task setting, called test-agnostic long-tailed recognition, where the training class distribution is long-tailed while the test class distribution is unknown and can be skewed arbitrarily.

23, TITLE: Attention-Guided NIR Image Colorization Via Adaptive Fusion of Semantic and Texture Clues
AUTHORS: Xingxing Yang ; Jie Chen ; Zaifeng Yang ; Zhenghua Chen
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this work, we propose a novel Attention-based NIR image colorization framework via Adaptive Fusion of Semantic and Texture clues, aiming at achieving these goals within the same framework.

24, TITLE: Towards Privacy-preserving Explanations in Medical Image Analysis
AUTHORS: H. Montenegro ; W. Silva ; J. S. Cardoso
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this work, we analyze existing privacy-preserving methods and their respective capacity to anonymize medical data while preserving disease-related semantic features.

25, TITLE: Critic Guided Segmentation of Rewarding Objects in First-Person Views
AUTHORS: ANDREW MELNIK et. al.
CATEGORY: cs.CV [cs.CV, cs.AI]
HIGHLIGHT: This work discusses a learning approach to mask rewarding objects in images using sparse reward signals from an imitation learning dataset.

26, TITLE: A Comparison of Supervised and Unsupervised Deep Learning Methods for Anomaly Detection in Images
AUTHORS: Vincent Wilmet ; Sauraj Verma ; Tabea Redl ; H�kon Sandaker ; Zhenning Li
CATEGORY: cs.CV [cs.CV, cs.LG, eess.IV]
HIGHLIGHT: Therefore, in this paper we investigate different methods of deep learning, including supervised and unsupervised learning, for anomaly detection applied to a quality assurance use case.

27, TITLE: Monocular Visual Analysis for Electronic Line Calling of Tennis Games
AUTHORS: Yuanzhou Chen ; Shaobo Cai ; Yuxin Wang ; Junchi Yan
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: We propose a monocular vision technology based ELC method.

28, TITLE: RankSRGAN: Super Resolution Generative Adversarial Networks with Learning to Rank
AUTHORS: Wenlong Zhang ; Yihao Liu ; Chao Dong ; Yu Qiao
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: To address the problem, we propose Super-Resolution Generative Adversarial Networks with Ranker (RankSRGAN) to optimize generator in the direction of different perceptual metrics.

29, TITLE: Image-Hashing-Based Anomaly Detection for Privacy-Preserving Online Proctoring
AUTHORS: Waheeb Yaqub ; Manoranjan Mohanty ; Basem Suleiman
CATEGORY: cs.CR [cs.CR, cs.CV, cs.HC]
HIGHLIGHT: In this paper, we propose a privacy-preserving online proctoring system.

30, TITLE: Characterizing Generalization Under Out-Of-Distribution Shifts in Deep Metric Learning
AUTHORS: TIMO MILBICH et. al.
CATEGORY: cs.LG [cs.LG, cs.CV]
HIGHLIGHT: In this work, we systematically construct train-test splits of increasing difficulty and present the ooDML benchmark to characterize generalization under out-of-distribution shifts in DML.

31, TITLE: Follow Your Path: A Progressive Method for Knowledge Distillation
AUTHORS: Wenxian Shi ; Yuxuan Song ; Hao Zhou ; Bohan Li ; Lei Li
CATEGORY: cs.LG [cs.LG, cs.CV]
HIGHLIGHT: In this paper, we propose ProKT, a new model-agnostic method by projecting the supervision signals of a teacher model into the student's parameter space.

32, TITLE: Generative Video Transformer: Can Objects Be The Words?
AUTHORS: Yi-Fu Wu ; Jaesik Yoon ; Sungjin Ahn
CATEGORY: cs.LG [cs.LG, cs.CL, cs.CV]
HIGHLIGHT: In this paper, we propose the Object-Centric Video Transformer (OCVT) which utilizes an object-centric approach for decomposing scenes into tokens suitable for use in a generative video transformer.

33, TITLE: FoleyGAN: Visually Guided Generative Adversarial Network-Based Synchronous Sound Generation in Silent Videos
AUTHORS: Sanchita Ghose ; John J. Prevost
CATEGORY: cs.LG [cs.LG, cs.AI, cs.CV, cs.MM, cs.SD, 68T10 (primary) 68T07, 68U10(secondary), I.5.4; I.2.10; J.5]
HIGHLIGHT: In this research we introduce a novel task of guiding a class conditioned generative adversarial network with the temporal visual information of a video input for visual to sound generation task adapting the synchronicity traits between audio-visual modalities.

34, TITLE: Confidence Aware Neural Networks for Skin Cancer Detection
AUTHORS: DONYA KHALEDYAN et. al.
CATEGORY: eess.IV [eess.IV, cs.CV, cs.LG]
HIGHLIGHT: To address this issue, this work presents three different methods for quantifying uncertainties for skin cancer detection from images.

35, TITLE: Protecting Semantic Segmentation Models By Using Block-wise Image Encryption with Secret Key from Unauthorized Access
AUTHORS: Hiroki Ito ; MaungMaung AprilPyone ; Hitoshi Kiya
CATEGORY: eess.IV [eess.IV, cs.CV, cs.LG]
HIGHLIGHT: In this paper, we propose to protect semantic segmentation models from unauthorized access by utilizing block-wise transformation with a secret key for the first time.

36, TITLE: OSLO: On-the-Sphere Learning for Omnidirectional Images and Its Application to 360-degree Image Compression
AUTHORS: Navid Mahmoudian Bidgoli ; Roberto G. de A. Azevedo ; Thomas Maugey ; Aline Roumy ; Pascal Frossard
CATEGORY: eess.IV [eess.IV, cs.CV]
HIGHLIGHT: In this paper, we study the learning of representation models for omnidirectional images and propose to use the properties of HEALPix uniform sampling of the sphere to redefine the mathematical tools used in deep learning models for omnidirectional images.

37, TITLE: SynthSeg: Domain Randomisation for Segmentation of Brain MRI Scans of Any Contrast and Resolution
AUTHORS: BENJAMIN BILLOT et. al.
CATEGORY: eess.IV [eess.IV, cs.CV]
HIGHLIGHT: We introduce SynthSeg, the first segmentation CNN agnostic to brain MRI scans of any contrast and resolution.

38, TITLE: Convolutional Module for Heart Localization and Segmentation in MRI
AUTHORS: Daniel Lima ; Catharine Graves ; Marco Gutierrez ; Bruno Brandoli ; Jose Rodrigues-Jr
CATEGORY: eess.IV [eess.IV, cs.AI, cs.CV, 68T07, 92C55, 92B20, I.2.10; I.4.6; I.5.1; J.3]
HIGHLIGHT: In this paper we describe Visual-Motion-Focus (VMF), a module that detects the heart motion in the 4D MRI sequence, and highlights ROIs by focusing a Radial Basis Function (RBF) on the estimated motion field.

39, TITLE: Quality and Complexity Assessment of Learning-Based Image Compression Solutions
AUTHORS: Jo�o Dick ; Brunno Abreu ; Mateus Grellert ; Sergio Bampi
CATEGORY: eess.IV [eess.IV, cs.CV]
HIGHLIGHT: This work presents an analysis of state-of-the-art learning-based image compression techniques.

40, TITLE: DeepSMILE: Self-supervised Heterogeneity-aware Multiple Instance Learning for DNA Damage Response Defect Classification Directly from H&E Whole-slide Images
AUTHORS: Yoni Schirris ; Efstratios Gavves ; Iris Nederlof ; Hugo Mark Horlings ; Jonas Teuwen
CATEGORY: eess.IV [eess.IV, cs.CV]
HIGHLIGHT: We propose a Deep learning-based weak label learning method for analysing whole slide images (WSIs) of Hematoxylin and Eosin (H&E) stained tumorcells not requiring pixel-level or tile-level annotations using Self-supervised pre-training and heterogeneity-aware deep Multiple Instance LEarning (DeepSMILE).

41, TITLE: A Review of Generative Adversarial Networks in Cancer Imaging: New Applications, New Solutions
AUTHORS: RICHARD OSUALA et. al.
CATEGORY: eess.IV [eess.IV, cs.CV, cs.LG]
HIGHLIGHT: In this review, we assess the potential of GANs to address a number of key challenges of cancer imaging, including data scarcity and imbalance, domain and dataset shifts, data access and privacy, data annotation and quantification, as well as cancer detection, tumour profiling and treatment planning.

42, TITLE: Automated Segmentation and Volume Measurement of Intracranial Carotid Artery Calcification on Non-Contrast CT
AUTHORS: GERDA BORTSOVA et. al.
CATEGORY: eess.IV [eess.IV, cs.CV]
HIGHLIGHT: Purpose: To evaluate a fully-automated deep-learning-based method for assessment of intracranial carotid artery calcification (ICAC).