计算机视觉论文-2021-09-15

本专栏是计算机视觉方向论文收集积累，时间：2021年9月15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: MindCraft: Theory of Mind Modeling for Situated Dialogue in Collaborative Tasks
AUTHORS: Cristian-Paul Bara ; Sky CH-Wang ; Joyce Chai
CATEGORY: cs.AI [cs.AI, cs.CL, cs.CV, cs.LG]
HIGHLIGHT: To enable theory of mind modeling in situated interactions, we introduce a fine-grained dataset of collaborative tasks performed by pairs of human subjects in the 3D virtual blocks world of Minecraft.

2, TITLE: Broaden The Vision: Geo-Diverse Visual Commonsense Reasoning
AUTHORS: Da Yin ; Liunian Harold Li ; Ziniu Hu ; Nanyun Peng ; Kai-Wei Chang
CATEGORY: cs.CL [cs.CL, cs.AI, cs.CV]
HIGHLIGHT: In this paper, we construct a Geo-Diverse Visual Commonsense Reasoning dataset (GD-VCR) to test vision-and-language models' ability to understand cultural and geo-location-specific commonsense.

3, TITLE: Dynamic Attentive Graph Learning for Image Restoration
AUTHORS: Chong Mou ; Jian Zhang ; Zhuoyuan Wu
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: To rectify these weaknesses, in this paper, we propose a dynamic attentive graph learning model (DAGL) to explore the dynamic non-local property on patch level for image restoration.

4, TITLE: Adaptive Proposal Generation Network for Temporal Sentence Localization in Videos
AUTHORS: Daizong Liu ; Xiaoye Qu ; Jianfeng Dong ; Pan Zhou
CATEGORY: cs.CV [cs.CV, cs.CL]
HIGHLIGHT: In this paper, we propose an Adaptive Proposal Generation Network (APGN) to maintain the segment-level interaction while speeding up the efficiency.

5, TITLE: AdaPruner: Adaptive Channel Pruning and Effective Weights Inheritance
AUTHORS: Xiangcheng Liu ; Jian Cao ; Hongyi Yao ; Wenyu Sun ; Yuan Zhang
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this paper, we propose a pruning framework that adaptively determines the number of each layer's channels as well as the wights inheritance criteria for sub-network.

6, TITLE: Identifying Partial Mouse Brain Microscopy Images from Allen Reference Atlas Using A Contrastively Learned Semantic Space
AUTHORS: Justinas Antanavicius ; Roberto Leiras Gonzalez ; Raghavendra Selvan
CATEGORY: cs.CV [cs.CV, cs.LG, stat.ML]
HIGHLIGHT: This work explores Siamese Networks as the method for finding corresponding 2D reference atlas plates for given partial 2D mouse brain images.

7, TITLE: Multi-Level Features Contrastive Networks for Unsupervised Domain Adaptation
AUTHORS: LE LIU et. al.
CATEGORY: cs.CV [cs.CV, cs.AI]
HIGHLIGHT: In this paper, we develop this work on the method of class-level alignment.

8, TITLE: Sensor Adversarial Traits: Analyzing Robustness of 3D Object Detection Sensor Fusion Models
AUTHORS: Won Park ; Nan Li ; Qi Alfred Chen ; Z. Morley Mao
CATEGORY: cs.CV [cs.CV, cs.LG]
HIGHLIGHT: In this work, we perform the first study to analyze the robustness of a high-performance, open source sensor fusion model architecture towards adversarial attacks and challenge the popular belief that the use of additional sensors automatically mitigate the risk of adversarial attacks.

9, TITLE: Progressively Guide to Attend: An Iterative Alignment Framework for Temporal Sentence Grounding
AUTHORS: Daizong Liu ; Xiaoye Qu ; Pan Zhou
CATEGORY: cs.CV [cs.CV, cs.CL]
HIGHLIGHT: In this paper, we propose an Iterative Alignment Network (IA-Net) for TSG task, which iteratively interacts inter- and intra-modal features within multiple steps for more accurate grounding.

10, TITLE: From Heatmaps to Structural Explanations of Image Classifiers
AUTHORS: LI FUXIN et. al.
CATEGORY: cs.CV [cs.CV, cs.LG]
HIGHLIGHT: In this paper, we attempt to share those insights and opinions with the readers with the hope that some of them will be informative for future researchers on explainable deep learning.

11, TITLE: POPCORN: Progressive Pseudo-labeling with Consistency Regularization and Neighboring
AUTHORS: REDA ABDELLAH KAMRAOUI et. al.
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this work, we propose POPCORN, a novel method combining consistency regularization and pseudo-labeling designed for image segmentation.

12, TITLE: Camera-Tracklet-Aware Contrastive Learning for Unsupervised Vehicle Re-Identification
AUTHORS: Jongmin Yu ; Junsik Kim ; Minkyung Kim ; Hyeontaek Oh
CATEGORY: cs.CV [cs.CV, cs.AI]
HIGHLIGHT: In this paper, we propose camera-tracklet-aware contrastive learning (CTACL) using the multi-camera tracklet information without vehicle identity labels.

13, TITLE: A Deep Learning Approach for Masking Fetal Gender in Ultrasound Images
AUTHORS: AMIT BORUNDIYA et. al.
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: This work proposes a deep learning object detection approach to accurately mask fetal gender in US images in order to increase the accessibility of the technology.

14, TITLE: High-Fidelity GAN Inversion for Image Attribute Editing
AUTHORS: Tengfei Wang ; Yong Zhang ; Yanbo Fan ; Jue Wang ; Qifeng Chen
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: We present a novel high-fidelity generative adversarial network (GAN) inversion framework that enables attribute editing with image-specific details well-preserved (e.g., background, appearance and illumination).

15, TITLE: Tesla-Rapture: A Lightweight Gesture Recognition System from MmWave Radar Point Clouds
AUTHORS: DARIUSH SALAMI et. al.
CATEGORY: cs.CV [cs.CV, cs.LG, eess.IV]
HIGHLIGHT: We present Tesla-Rapture, a gesture recognition interface for point clouds generated by mmWave Radars.

16, TITLE: MotionHint: Self-Supervised Monocular Visual Odometrywith Motion Constraints
AUTHORS: Cong Wang ; Yu-Ping Wang ; Dinesh Manocha
CATEGORY: cs.CV [cs.CV, cs.RO]
HIGHLIGHT: We present a novel self-supervised algorithmnamedMotionHintfor monocular visual odometry (VO) that takes motion constraints into account.

17, TITLE: Image-Based Alignment of 3D Scans
AUTHORS: Dolores Messer ; Jakob Wilm ; Eythor R. Eiriksson ; Vedrana A. Dahl ; Anders B. Dahl
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this paper, we propose a fully automatic method for aligning the scans of an object in two different poses.

18, TITLE: A Semantic Indexing Structure for Image Retrieval
AUTHORS: YING WANG et. al.
CATEGORY: cs.CV [cs.CV, cs.IR, 68T20, H.3.1]
HIGHLIGHT: In this paper a new classification-based indexing structure, called Semantic Indexing Structure (SIS), is proposed, in which we utilize the semantic categories rather than clustering centers to create database partitions, such that the proposed index SIS can be combined with feature extractors without the restriction of dimensions.

19, TITLE: Luminance Attentive Networks for HDR Image and Panorama Reconstruction
AUTHORS: HANNING YU et. al.
CATEGORY: cs.CV [cs.CV, eess.IV]
HIGHLIGHT: This paper proposes a luminance attentive network named LANet for HDR reconstruction from a single LDR image.

20, TITLE: One-Class Meta-Learning: Towards Generalizable Few-Shot Open-Set Classification
AUTHORS: Jedrzej Kozerawski ; Matthew Turk
CATEGORY: cs.CV [cs.CV, cs.LG]
HIGHLIGHT: We introduce two independent few-shot one-class classification methods: Meta Binary Cross-Entropy (Meta-BCE), which learns a separate feature representation for one-class classification, and One-Class Meta-Learning (OCML), which learns to generate one-class classifiers given standard multiclass feature representation.

21, TITLE: Sampling Network Guided Cross-Entropy Method for Unsupervised Point Cloud Registration
AUTHORS: HAOBO JIANG et. al.
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this paper, by modeling the point cloud registration task as a Markov decision process, we propose an end-to-end deep model embedded with the cross-entropy method (CEM) for unsupervised 3D registration.

22, TITLE: LRWR: Large-Scale Benchmark for Lip Reading in Russian Language
AUTHORS: Evgeniy Egorov ; Vasily Kostyumov ; Mikhail Konyk ; Sergey Kolesnikov
CATEGORY: cs.CV [cs.CV, cs.LG]
HIGHLIGHT: In this paper, we introduce a naturally distributed large-scale benchmark for lipreading in Russian language, named LRWR, which contains 235 classes and 135 speakers.

23, TITLE: Spiking Neural Networks for Visual Place Recognition Via Weighted Neuronal Assignments
AUTHORS: Somayeh Hussaini ; Michael Milford ; Tobias Fischer
CATEGORY: cs.CV [cs.CV, cs.RO]
HIGHLIGHT: This research introduces the first high performance SNN for the Visual Place Recognition (VPR) task: given a query image, the SNN has to find the closest match out of a list of reference images.

24, TITLE: Improved Few-shot Segmentation By Redifinition of The Roles of Multi-level CNN Features
AUTHORS: Zhijie Wang ; Masanori Suganuma ; Takayuki Okatani
CATEGORY: cs.CV [cs.CV, cs.AI]
HIGHLIGHT: In this paper, we reinterpret this widely employed approach by redifining the roles of the multi-level features; we swap the primary and secondary roles.

25, TITLE: Cross-Region Domain Adaptation for Class-level Alignment
AUTHORS: Zhijie Wang ; Xing Liu ; Masanori Suganuma ; Takayuki Okatani
CATEGORY: cs.CV [cs.CV, cs.AI]
HIGHLIGHT: To cope with this, we propose a method that applies adversarial training to align two feature distributions in the target domain.

26, TITLE: High-Resolution Image Harmonization Via Collaborative Dual Transformations
AUTHORS: WENYAN CONG et. al.
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this work, we propose a high-resolution image harmonization network with Collaborative Dual Transformation (CDTNet) to combine pixel-to-pixel transformation and RGB-to-RGB transformation coherently in an end-to-end framework.

27, TITLE: Incremental Abstraction in Distributed Probabilistic SLAM Graphs
AUTHORS: Joseph Ortiz ; Talfan Evans ; Edgar Sucar ; Andrew J. Davison
CATEGORY: cs.CV [cs.CV, cs.RO]
HIGHLIGHT: We present a distributed, graph-based SLAM framework for incrementally building scene graphs based on two novel components.

28, TITLE: Dodging Attack Using Carefully Crafted Natural Makeup
AUTHORS: Nitzan Guetta ; Asaf Shabtai ; Inderjeet Singh ; Satoru Momiyama ; Yuval Elovici
CATEGORY: cs.CV [cs.CV, cs.CR, cs.LG]
HIGHLIGHT: In this study, we present a novel black-box AML attack which carefully crafts natural makeup, which, when applied on a human participant, prevents the participant from being identified by facial recognition models.

29, TITLE: Space Time Recurrent Memory Network
AUTHORS: Hung Nguyen ; Fuxin Li
CATEGORY: cs.CV [cs.CV, cs.AI]
HIGHLIGHT: We propose a novel visual memory network architecture for the learning and inference problem in the spatial-temporal domain.

30, TITLE: Monocular Camera Localization for Automated Vehicles Using Image Retrieval
AUTHORS: Eunhyek Joa ; Francesco Borrelli
CATEGORY: cs.CV [cs.CV, cs.SY, eess.SY]
HIGHLIGHT: We address the problem of finding the current position and heading angle of an autonomous vehicle in real-time using a single camera.

31, TITLE: Anomaly Attribution of Multivariate Time Series Using Counterfactual Reasoning
AUTHORS: Violeta Teodora Trifunov ; Maha Shadaydeh ; Bj�rn Barz ; Joachim Denzler
CATEGORY: cs.LG [cs.LG, cs.CV, stat.ML]
HIGHLIGHT: We aim to answer the counterfactual question of would the anomalous event have occurred if the subset of the involved variables had been more similarly distributed to the data outside of the anomalous interval.

32, TITLE: Automatic Hippocampal Surface Generation Via 3D U-net and Active Shape Modeling with Hybrid Particle Swarm Optimization
AUTHORS: Pinyuan Zhong ; Yue Zhang ; Xiaoying Tang
CATEGORY: cs.NE [cs.NE, cs.CV, cs.LG, eess.IV]
HIGHLIGHT: In this paper, we proposed and validated a fully automatic pipeline for hippocampal surface generation via 3D U-net coupled with active shape modeling (ASM).

33, TITLE: ImUnity: A Generalizable VAE-GAN Solution for Multicenter MR Image Harmonization
AUTHORS: Stenzel Cackowski ; Emmanuel L. Barbier ; Michel Dojat ; Thomas Christen
CATEGORY: eess.IV [eess.IV, cs.CV, cs.LG]
HIGHLIGHT: Using 3 open source databases (ABIDE, OASIS and SRPBS), which contain MR images from multiple acquisition scanner types or vendors and a large range of subjects ages, we show that ImUnity: (1) outperforms state-of-the-art methods in terms of quality of images generated using traveling subjects; (2) removes sites or scanner biases while improving patients classification; (3) harmonizes data coming from new sites or scanners without the need for an additional fine-tuning and (4) allows the selection of multiple MR reconstructed images according to the desired applications.

34, TITLE: Multi-Scale Input Strategies for Medulloblastoma Tumor Classification Using Deep Transfer Learning
AUTHORS: Marcel Bengs ; Satish Pant ; Michael Bockmayr ; Ulrich Sch�ller ; Alexander Schlaefer
CATEGORY: eess.IV [eess.IV, cs.CV]
HIGHLIGHT: We study the impact of tile size and input strategy and classify the two major histopathological subtypes-Classic and Demoplastic/Nodular.

35, TITLE: Dense Deep Unfolding Network with 3D-CNN Prior for Snapshot Compressive Imaging
AUTHORS: Zhuoyuan Wu ; Jian Zhang ; Chong Mou
CATEGORY: eess.IV [eess.IV, cs.CV]
HIGHLIGHT: For the sake of building a fast and accurate SCI recovery algorithm, we incorporate the interpretability of model-based methods and the speed of learning-based ones and present a novel dense deep unfolding network (DUN) with 3D-CNN prior for SCI, where each phase is unrolled from an iteration of Half-Quadratic Splitting (HQS).

36, TITLE: Physics Driven Domain Specific Transporter Framework with Attention Mechanism for Ultrasound Imaging
AUTHORS: ARPAN TRIPATHI et. al.
CATEGORY: eess.IV [eess.IV, cs.CV, cs.LG]
HIGHLIGHT: In this paper, we propose an unsupervised, physics driven domain specific transporter framework with an attention mechanism to identify relevant key points with applications in ultrasound imaging.

37, TITLE: 3-Dimensional Deep Learning with Spatial Erasing for Unsupervised Anomaly Segmentation in Brain MRI
AUTHORS: Marcel Bengs ; Finn Behrendt ; Julia Kr�ger ; Roland Opfer ; Alexander Schlaefer
CATEGORY: eess.IV [eess.IV, cs.CV]
HIGHLIGHT: We propose 3D deep learning methods for UAD in brain MRI combined with 3D erasing and demonstrate that 3D methods clearly outperform their 2D counterpart for anomaly segmentation.

38, TITLE: Cross-Modality Domain Adaptation for Vestibular Schwannoma and Cochlea Segmentation
AUTHORS: HAN LIU et. al.
CATEGORY: eess.IV [eess.IV, cs.CV]
HIGHLIGHT: In this work, we aim to tackle the VS and cochlea segmentation problem in an unsupervised domain adaptation setting.

39, TITLE: COVID-Net MLSys: Designing COVID-Net for The Clinical Workflow
AUTHORS: AUDREY G. CHUNG et. al.
CATEGORY: eess.IV [eess.IV, cs.CV]
HIGHLIGHT: In this study, we take a machine learning and systems (MLSys) perspective to design a system for COVID-19 patient screening with the clinical workflow in mind.