计算机视觉论文-2021-03-01

本专栏是计算机视觉方向论文收集积累，时间：2021年3月1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: DeepSZ: Identification of Sunyaev-Zel'dovich Galaxy Clusters Using Deep Learning
AUTHORS: ZHEN LIN et. al.
CATEGORY: astro-ph.CO [astro-ph.CO, cs.CV, cs.LG]
HIGHLIGHT: We present a comparison between two methods of cluster identification: the standard Matched Filter (MF) method in SZ cluster finding and a method using Convolutional Neural Networks (CNN).

2, TITLE: Natural Language Video Localization: A Revisit in Span-based Question Answering Framework
AUTHORS: HAO ZHANG et. al.
CATEGORY: cs.CL [cs.CL, cs.CV]
HIGHLIGHT: In this work, we address the NLVL from a new perspective, i.e., span-based question answering (QA), by treating the input video as a text passage.

3, TITLE: MixSearch: Searching for Domain Generalized Medical Image Segmentation Architectures
AUTHORS: LUYAN LIU et. al.
CATEGORY: cs.CV [cs.CV, cs.AI]
HIGHLIGHT: In this paper, we address this point by proposing to search for generalizable U-shape architectures on a composited dataset that mixes medical images from multiple segmentation tasks and domains creatively, which is named MixSearch.

4, TITLE: A Universal Model for Cross Modality Mapping By Relational Reasoning
AUTHORS: ZUN LI et. al.
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: Existing methods usually formulate the mapping function as the similarity measure between the pair of instance features, which are embedded to a common space.

5, TITLE: Dual-MTGAN: Stochastic and Deterministic Motion Transfer for Image-to-Video Synthesis
AUTHORS: Fu-En Yang ; Jing-Cheng Chang ; Yuan-Hao Lee ; Yu-Chiang Frank Wang
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this paper, we propose Dual Motion Transfer GAN (Dual-MTGAN), which takes image and video data as inputs while learning disentangled content and motion representations.

6, TITLE: Mitigating Domain Mismatch in Face Recognition Using Style Matching
AUTHORS: Chun-Hsien Lin ; Bing-Fei Wu
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this paper, we formulate domain mismatch in face recognition as a style mismatch problem for which we propose two methods.

7, TITLE: Zero-Shot Learning Based on Knowledge Sharing
AUTHORS: Zeng Ting ; Xiang Hongxin ; Xie Cheng ; Yang Yun ; Liu Qing
CATEGORY: cs.CV [cs.CV, cs.AI]
HIGHLIGHT: In this paper, we introduce knowledge sharing (KS) to enrich the representation of semantic features.

8, TITLE: Multi-Domain Learning By Meta-Learning: Taking Optimal Steps in Multi-Domain Loss Landscapes By Inner-Loop Learning
AUTHORS: ANTHONY SICILIA et. al.
CATEGORY: cs.CV [cs.CV, cs.LG, eess.IV]
HIGHLIGHT: In this paper, given emerging multi-modal data (e.g., various structural neuroimaging modalities), we aim to enable MDL purely algorithmically so that widely used neural networks can trivially achieve MDL in a model-independent manner.

9, TITLE: Robust Pollen Imagery Classification with Generative Modeling and Mixup Training
AUTHORS: Jaideep Murkute
CATEGORY: cs.CV [cs.CV, cs.LG, eess.IV]
HIGHLIGHT: In this work, we present and a robust deep learning framework that can generalize well for pollen grain aerobiological imagery classification.

10, TITLE: A Reconfigurable Winograd CNN Accelerator with Nesting Decomposition Algorithm for Computing Convolution with Large Filters
AUTHORS: Jingbo Jiang ; Xizi Chen ; Chi-Ying Tsui
CATEGORY: cs.CV [cs.CV, cs.AI]
HIGHLIGHT: This work proposes a nested Winograd algorithm to iteratively decompose a large filter into a sequence of 3x3 tiles which can then be accelerated with a 3x3 Winograd algorithm.

11, TITLE: Where to Look at The Movies : Analyzing Visual Attention to Understand Movie Editing
AUTHORS: Alexandre Bruckert ; Marc Christie ; Olivier Le Meur
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this work, we expose strong links between movie editing and spectators scanpaths, and open several leads on how the knowledge of editing information could improve human visual attention modeling for cinematic content. In order to provide a quantitative analysis of the relationship between those tools and gaze patterns, we propose a new eye-tracking database, containing gaze pattern information on movie sequences, as well as editing annotations, and we show how state-of-the-art computational saliency techniques behave on this dataset.

12, TITLE: Nested-block Self-attention for Robust Radiotherapy Planning Segmentation
AUTHORS: HARINI VEERARAGHAVAN et. al.
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In order to address these challenges, we developed a computationally efficient nested block self-attention (NBSA) method that can be combined with any convolutional network.

13, TITLE: Continuous Face Aging Generative Adversarial Networks
AUTHORS: Seogkyu Jeon ; Pilhyeon Lee ; Kibeom Hong ; Hyeran Byun
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: To this end, we propose the continuous face aging generative adversarial networks (CFA-GAN).

14, TITLE: Domain Adapting Ability of Self-Supervised Learning for Face Recognition
AUTHORS: Chun-Hsien Lin ; Bing-Fei Wu
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this paper, self-supervised learning is adopted to learn a better embedding space where the subjects in target domain are more distinguishable.

15, TITLE: Unifying Remote Sensing Image Retrieval and Classification with Robust Fine-tuning
AUTHORS: Dimitri Gominski ; Val�rie Gouet-Brunet ; Liming Chen
CATEGORY: cs.CV [cs.CV, cs.IR]
HIGHLIGHT: We aim at unifying remote sensing image retrieval and classification with a new large-scale training and testing dataset, SF300, including both vertical and oblique aerial images and made available to the research community, and an associated fine-tuning method.

16, TITLE: Point Cloud Upsampling and Normal Estimation Using Deep Learning for Robust Surface Reconstruction
AUTHORS: Rajat Sharma ; Tobias Schwandt ; Christian Kunert ; Steffen Urban ; Wolfgang Broll
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this paper, we present a novel deep learning architecture for point cloud upsampling that enables subsequent stable and smooth surface reconstruction.

17, TITLE: Knowledge Distillation Circumvents Nonlinearity for Optical Convolutional Neural Networks
AUTHORS: Jinlin Xiang ; Shane Colburn ; Arka Majumdar ; Eli Shlizerman
CATEGORY: cs.CV [cs.CV, cs.ET, cs.LG]
HIGHLIGHT: Here, we propose a Spectral CNN Linear Counterpart (SCLC) network architecture and develop a Knowledge Distillation (KD) approach to circumvent the need for a nonlinearity and successfully train such networks.

18, TITLE: Using Deep Learning to Automate The Detection of Flaws in Nuclear Fuel Channel UT Scans
AUTHORS: Issam Hammad ; Ryan Simpson ; Hippolyte Djonon Tsague ; Sarah Hall
CATEGORY: cs.CV [cs.CV, eess.IV]
HIGHLIGHT: In this paper, a proof of concept (PoC) that automates the detection of flaws in nuclear fuel channel UT scans using a convolutional neural network (CNN) is presented.

19, TITLE: Class Knowledge Overlay to Visual Feature Learning for Zero-Shot Image Classification
AUTHORS: CHENG XIE et. al.
CATEGORY: cs.CV [cs.CV, cs.AI]
HIGHLIGHT: In this paper, we propose a novel zero-shot learning approach, GAN-CST, based on class knowledge to visual feature learning to tackle the problem.

20, TITLE: Many-to-One Distribution Learning and K-Nearest Neighbor Smoothing for Thoracic Disease Identification
AUTHORS: Yi Zhou ; Lei Huang ; Tianfei Zhou ; Ling Shao
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this paper, we propose many-to-one distribution learning (MODL) and K-nearest neighbor smoothing (KNNS) methods from two perspectives to improve a single model's disease identification performance, rather than focusing on an ensemble of models.

21, TITLE: Improving Robustness of Learning-based Autonomous Steering Using Adversarial Images
AUTHORS: YU SHEN et. al.
CATEGORY: cs.CV [cs.CV, cs.LG, cs.RO]
HIGHLIGHT: In this work, we address this critical issue by introducing a framework for analyzing robustness of the learning algorithm w.r.t varying quality in the image input for autonomous driving.

22, TITLE: Boundary-induced and Scene-aggregated Network for Monocular Depth Prediction
AUTHORS: FENG XUE et. al.
CATEGORY: cs.CV [cs.CV, cs.AI]
HIGHLIGHT: To tackle these two issues, we propose the Boundary-induced and Scene-aggregated network (BS-Net).

23, TITLE: ACDnet: An Action Detection Network for Real-time Edge Computing Based on Flow-guided Feature Approximation and Memory Aggregation
AUTHORS: Yu Liu ; Fan Yang ; Dominique Ginhac
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this paper, we propose ACDnet, a compact action detection network targeting real-time edge computing which addresses both efficiency and accuracy.

24, TITLE: Surgical Visual Domain Adaptation: Results from The MICCAI 2020 SurgVisDom Challenge
AUTHORS: ANEEQ ZIA et. al.
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In particular, we propose to use video from virtual reality (VR) simulations of surgical exercises in robotic-assisted surgery to develop algorithms to recognize tasks in a clinical-like setting. We also release the challenge dataset publicly at https://www.synapse.org/surgvisdom2020.

25, TITLE: Accurate Visual-Inertial SLAM By Feature Re-identification
AUTHORS: Xiongfeng Peng ; Zhihua Liu ; Qiang Wang ; Yun-Tae Kim ; Myungjae Jeon
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: We propose a novel feature re-identification method for real-time visual-inertial SLAM.

26, TITLE: Machine Biometrics -- Towards Identifying Machines in A Smart City Environment
AUTHORS: G. K. Sidiropoulos ; G. A. Papakostas
CATEGORY: cs.LG [cs.LG, cs.AI, cs.CR, cs.CV, I.5.4]
HIGHLIGHT: The concept of machine biometrics is proposed in this work for the first time, as a way to authenticate machine identities interacting with humans in everyday life.

27, TITLE: Nonlinear Projection Based Gradient Estimation for Query Efficient Blackbox Attacks
AUTHORS: HUICHEN LI et. al.
CATEGORY: cs.LG [cs.LG, cs.CV]
HIGHLIGHT: We aim to bridge the gap between the two by investigating how to efficiently estimate gradient based on a projected low-dimensional space.

28, TITLE: What Doesn't Kill You Makes You Robust(er): Adversarial Training Against Poisons and Backdoors
AUTHORS: JONAS GEIPING et. al.
CATEGORY: cs.LG [cs.LG, cs.CR, cs.CV]
HIGHLIGHT: In this work, we extend the adversarial training framework to instead defend against (training-time) poisoning and backdoor attacks.

29, TITLE: Panoramic Annular SLAM with Loop Closure and Global Optimization
AUTHORS: Hao Chen ; Weijian Hu ; Kailun Yang ; Jian Bai ; Kaiwei Wang
CATEGORY: cs.RO [cs.RO, cs.CV, eess.IV]
HIGHLIGHT: In this paper, we propose PA-SLAM, a monocular panoramic annular visual SLAM system with loop closure and global optimization.

30, TITLE: 3D Vessel Reconstruction in OCT-Angiography Via Depth Map Estimation
AUTHORS: SHUAI YU et. al.
CATEGORY: eess.IV [eess.IV, cs.CV, cs.LG]
HIGHLIGHT: In this paper, we introduce a novel 3D vessel reconstruction framework based on the estimation of vessel depth maps from OCTA images.

31, TITLE: Texture-aware Video Frame Interpolation
AUTHORS: Duolikun Danier ; David Bull
CATEGORY: eess.IV [eess.IV, cs.CV]
HIGHLIGHT: In this work, we study the impact of video textures on video frame interpolation, and propose a novel framework where, given an interpolation algorithm, separate models are trained on different textures.

32, TITLE: Robust Rational Polynomial Camera Modelling for SAR and Pushbroom Imaging
AUTHORS: Roland Akiki ; Roger Mar� ; Carlo de Franchis ; Jean-Michel Morel ; Gabriele Facciolo
CATEGORY: eess.IV [eess.IV, cs.CV]
HIGHLIGHT: This article describes a terrain-independent algorithm to accurately derive a RPC model from a set of 3D-2D point correspondences based on a regularized least squares fit.

33, TITLE: Convolution-Free Medical Image Segmentation Using Transformers
AUTHORS: Davood Karimi ; Serge Vasylechko ; Ali Gholipour
CATEGORY: eess.IV [eess.IV, cs.CV]
HIGHLIGHT: In this work we show that a different method, based entirely on self-attention between neighboring image patches and without any convolution operations, can achieve competitive or better results.