计算机视觉论文-2021-09-06

本专栏是计算机视觉方向论文收集积累，时间：2021年9月6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: CX-ToM: Counterfactual Explanations with Theory-of-Mind for Enhancing Human Trust in Image Recognition Models
AUTHORS: ARJUN R. AKULA et. al.
CATEGORY: cs.AI [cs.AI, cs.CV, cs.LG]
HIGHLIGHT: We propose CX-ToM, short for counterfactual explanations with theory-of mind, a new explainable AI (XAI) framework for explaining decisions made by a deep convolutional neural network (CNN).

2, TITLE: Remote Multilinear Compressive Learning with Adaptive Compression
AUTHORS: Dat Thanh Tran ; Moncef Gabbouj ; Alexandros Iosifidis
CATEGORY: cs.CV [cs.CV, eess.IV]
HIGHLIGHT: In this paper, we propose a novel optimization scheme that enables such a feature for MCL models.

3, TITLE: Roadscene2vec: A Tool for Extracting and Embedding Road Scene-Graphs
AUTHORS: ARNAV VAIBHAV MALAWADE et. al.
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: To enable the exploration of applications of road scene-graph representations, we introduce roadscene2vec: an open-source tool for extracting and embedding road scene-graphs.

4, TITLE: Safety-aware Motion Prediction with Unseen Vehicles for Autonomous Driving
AUTHORS: Xuanchi Ren ; Tao Yang ; Li Erran Li ; Alexandre Alahi ; Qifeng Chen
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this paper, we study a new task, safety-aware motion prediction with unseen vehicles for autonomous driving.

5, TITLE: DeepTracks: Geopositioning Maritime Vehicles in Video Acquired from A Moving Platform
AUTHORS: Jianli Wei ; Guanyu Xu ; Alper Yilmaz
CATEGORY: cs.CV [cs.CV, eess.IV]
HIGHLIGHT: The problem can be stated as follows: given imagery from a camera mounted on a moving platform with known GPS location as the only valid sensor, we predict the geoposition of a target boat visible in images.

6, TITLE: Ghost Loss to Question The Reliability of Training Data
AUTHORS: Adrien Deli�ge ; Anthony Cioppa ; Marc Van Droogenbroeck
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this work, we question the reliability of the annotated datasets.

7, TITLE: Wildfire Smoke Plume Segmentation Using Geostationary Satellite Imagery
AUTHORS: Jeff Wen ; Marshall Burke
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: This work uses deep convolutional neural networks to segment smoke plumes from geostationary satellite imagery.

8, TITLE: Neural Human Deformation Transfer
AUTHORS: Jean Basset ; Adnane Boukhayma ; Stefanie Wuhrer ; Franck Multon ; Edmond Boyer
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this work, we take a different approach and transform the identity of a character into a new identity without modifying the character's pose.

9, TITLE: 3D Human Shape Style Transfer
AUTHORS: Joao Regateiro ; Edmond Boyer
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this paper, we explore an alternative approach that transfers the source shape style onto the moving character.

10, TITLE: Representing Shape Collections with Alignment-Aware Linear Models
AUTHORS: Romain Loiseau ; Tom Monnier ; Lo�c Landrieu ; Mathieu Aubry
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this paper, we revisit the classical representation of 3D point clouds as linear shape models.

11, TITLE: Optimal Target Shape for LiDAR Pose Estimation
AUTHORS: Jiunn-Kai Huang ; William Clark ; Jessy W. Grizzle
CATEGORY: cs.CV [cs.CV, cs.RO]
HIGHLIGHT: This paper introduces the concept of optimizing target shape to remove pose ambiguity for LiDAR point clouds.

12, TITLE: Using Topological Framework for The Design of Activation Function and Model Pruning in Deep Neural Networks
AUTHORS: Yogesh Kochar ; Sunil Kumar Vengalil ; Neelam Sinha
CATEGORY: cs.CV [cs.CV, cs.CG, cs.LG]
HIGHLIGHT: Two independent contributions of this paper are 1) Novel activation function for faster training convergence 2) Systematic pruning of filters of models trained irrespective of activation function.

13, TITLE: Occlusion-Invariant Rotation-Equivariant Semi-Supervised Depth Based Cross-View Gait Pose Estimation
AUTHORS: XIAO GU et. al.
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: To address this, we propose a novel approach for cross-view generalization with an occlusion-invariant semi-supervised learning framework built upon a novel rotation-equivariant backbone.

14, TITLE: Access Control Using Spatially Invariant Permutation of Feature Maps for Semantic Segmentation Models
AUTHORS: Hiroki Ito ; MaungMaung AprilPyone ; Hitoshi Kiya
CATEGORY: cs.CV [cs.CV, cs.LG]
HIGHLIGHT: In this paper, we propose an access control method that uses the spatially invariant permutation of feature maps with a secret key for protecting semantic segmentation models.

15, TITLE: Deep Metric Learning for Ground Images
AUTHORS: Raaghav Radhakrishnan ; Jan Fabian Schmid ; Randolf Scholz ; Lars Schmidt-Thieme
CATEGORY: cs.CV [cs.CV, cs.RO]
HIGHLIGHT: For this purpose, we propose a deep metric learning approach that retrieves the most similar reference images to the query image.

16, TITLE: Information Symmetry Matters: A Modal-Alternating Propagation Network for Few-Shot Learning
AUTHORS: Zhong Ji ; Zhishen Hou ; Xiyao Liu ; Yanwei Pang ; Jungong Han
CATEGORY: cs.CV [cs.CV, cs.AI, cs.CL]
HIGHLIGHT: To address this problem, we propose a Modal-Alternating Propagation Network (MAP-Net) to supplement the absent semantic information of unlabeled samples, which builds information symmetry among all samples in both visual and semantic modalities.

17, TITLE: Ordinal Pooling
AUTHORS: Adrien Deli�ge ; Maxime Istasse ; Ashwani Kumar ; Christophe De Vleeschouwer ; Marc Van Droogenbroeck
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: To address this issue, a novel pooling scheme, named\emph{ ordinal pooling}, is introduced in this work.

18, TITLE: CAP-Net: Correspondence-Aware Point-view Fusion Network for 3D Shape Analysis
AUTHORS: Xinwei He ; Silin Cheng ; Song Bai ; Xiang Bai
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: To investigate this, we propose a novel Correspondence-Aware Point-view Fusion Net (CAPNet).

19, TITLE: Model-Based Parameter Optimization for Ground Texture Based Localization Methods
AUTHORS: Jan Fabian Schmid ; Stephan F. Simon ; Rudolf Mester
CATEGORY: cs.CV [cs.CV, cs.RO]
HIGHLIGHT: We tackle the issue of efficient parametrization of such methods, deriving a prediction model for localization performance, which requires only a small collection of sample images of an application area.

20, TITLE: Semantic Segmentation on VSPW Dataset Through Aggregation of Transformer Models
AUTHORS: Zixuan Chen ; Junhong Zou ; Xiaotao Wang
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this report, we briefly introduce the solutions of team 'BetterThing' for the ICCV2021 - Video Scene Parsing in the Wild Challenge.

21, TITLE: Deep Learning for Fitness
AUTHORS: Mahendran N
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: We present Fitness tutor, an application for maintaining correct posture during workout exercises or doing yoga.

22, TITLE: Segmentation of Turbulent Computational Fluid Dynamics Simulations with Unsupervised Ensemble Learning
AUTHORS: Maarja Bussov ; Joonas N�ttil�
CATEGORY: cs.CV [cs.CV, astro-ph.HE, cs.LG, physics.plasm-ph]
HIGHLIGHT: We apply the SCE algorithm to 2-dimensional simulation data snapshots of magnetically-dominated fully-kinetic turbulent plasma flows where accurate ROI boundaries are needed for geometrical measurements of intermittent flow structures known as current sheets.

23, TITLE: Video Pose Distillation for Few-Shot, Fine-Grained Sports Action Recognition
AUTHORS: James Hong ; Matthew Fisher ; Micha�l Gharbi ; Kayvon Fatahalian
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: We introduce Video Pose Distillation (VPD), a weakly-supervised technique to learn features for new video domains, such as individual sports that challenge pose estimation.

24, TITLE: Towards Learning Spatially Discriminative Feature Representations
AUTHORS: CHAOFEI WANG et. al.
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: We propose a novel loss function, termed as CAM-loss, to constrain the embedded feature maps with the class activation maps (CAMs) which indicate the spatially discriminative regions of an image for particular categories.

25, TITLE: Self-Taught Cross-Domain Few-Shot Learning with Weakly Supervised Object Localization and Task-Decomposition
AUTHORS: Xiyao Liu ; Zhong Ji ; Yanwei Pang ; Zhongfei Zhang
CATEGORY: cs.CV [cs.CV, cs.AI]
HIGHLIGHT: To this end, we propose a task-expansion-decomposition framework for CD-FSL, called Self-Taught (ST) approach, which alleviates the problem of non-target guidance by constructing task-oriented metric spaces.

26, TITLE: MitoVis: A Visually-guided Interactive Intelligent System for Neuronal Mitochondria Analysis
AUTHORS: JunYoung Choi ; Hakjun Lee ; Suyeon Kim ; Seok-Kyu Kwon ; Won-Ki Jeong
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: To address these issues, we introduce MitoVis, a novel visualization system for end-to-end data processing and interactive analysis of the morphology of neuronal mitochondria.

27, TITLE: Spatially Varying White Balancing for Mixed and Non-uniform Illuminants
AUTHORS: Teruaki Akazawa ; Yuma Kinoshita ; Hitoshi Kiya
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this paper, we propose a novel white balance adjustment, called "spatially varying white balancing," for single, mixed, and non-uniform illuminants.

28, TITLE: UnDeepLIO: Unsupervised Deep Lidar-Inertial Odometry
AUTHORS: Yiming Tu ; Jin Xie
CATEGORY: cs.CV [cs.CV, cs.RO]
HIGHLIGHT: In this paper, we design a novel framework for unsupervised lidar odometry with the IMU, which is never used in other deep methods.

29, TITLE: Dual-Camera Super-Resolution with Aligned Attention Modules
AUTHORS: Tengfei Wang ; Jiaxin Xie ; Wenxiu Sun ; Qiong Yan ; Qifeng Chen
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: We present a novel approach to reference-based super-resolution (RefSR) with the focus on dual-camera super-resolution (DCSR), which utilizes reference images for high-quality and high-fidelity results. We further explore the dual-camera super-resolution that is one promising application of RefSR, and build a dataset that consists of 146 image pairs from the main and telephoto cameras in a smartphone.

30, TITLE: A Reliable, Self-Adaptive Face Identification Framework Via Lyapunov Optimization
AUTHORS: Dohyeon Kim ; Joongheon Kim ; Jae young Bang
CATEGORY: cs.DC [cs.DC, cs.CV]
HIGHLIGHT: This paper proposes a novel, queue-aware FID framework that adapts the sampling rate to maximize the FID performance while avoiding a queue overflow by implementing the Lyapunov optimization.

31, TITLE: Edge-featured Graph Neural Architecture Search
AUTHORS: Shaofei Cai ; Liang Li ; Xinzhe Han ; Zheng-jun Zha ; Qingming Huang
CATEGORY: cs.LG [cs.LG, cs.AI, cs.CV]
HIGHLIGHT: To solve this problem, we incorporate edge features into graph search space and propose Edge-featured Graph Neural Architecture Search to find the optimal GNN architecture.

32, TITLE: Two Shifts for Crop Mapping: Leveraging Aggregate Crop Statistics to Improve Satellite-based Maps in New Regions
AUTHORS: Dan M. Kluger ; Sherrie Wang ; David B. Lobell
CATEGORY: stat.AP [stat.AP, cs.CV, cs.LG, 62P12 (primary) 62H30, I.4.m; I.m; J.2]
HIGHLIGHT: To adjust for shifts in features we propose a method to estimate and remove linear shifts in the mean feature vector.

33, TITLE: Deep Learning Approach for Hyperspectral Image Demosaicking, Spectral Correction and High-resolution RGB Reconstruction
AUTHORS: PEICHAO LI et. al.
CATEGORY: eess.IV [eess.IV, cs.CV]
HIGHLIGHT: In this work, we propose a deep learning-based image demosaicking algorithm for snapshot hyperspectral images using supervised learning methods.

34, TITLE: Automatic Foot Ulcer Segmentation Using An Ensemble of Convolutional Neural Networks
AUTHORS: Amirreza Mahbod ; Rupert Ecker ; Isabella Ellinger
CATEGORY: eess.IV [eess.IV, cs.CV]
HIGHLIGHT: In this work, we proposed an ensemble approach based on two encoder-decoder-based CNN models, namely LinkNet and UNet, to perform foot ulcer segmentation.

35, TITLE: Multi-centred Strong Augmentation Via Contrastive Learning for Unsupervised Lesion Detection and Segmentation
AUTHORS: YU TIAN et. al.
CATEGORY: eess.IV [eess.IV, cs.CV]
HIGHLIGHT: To address this challenge, we propose a novel self-supervised UAD pre-training algorithm, named Multi-centred Strong Augmentation via Contrastive Learning (MSACL).

36, TITLE: Unsupervised Multi-latent Space Reinforcement Learning Framework for Video Summarization in Ultrasound Imaging
AUTHORS: ROSHAN P MATHEWS et. al.
CATEGORY: eess.IV [eess.IV, cs.CV]
HIGHLIGHT: We propose a new unsupervised reinforcement learning (RL) framework with novel rewards that facilitates unsupervised learning avoiding tedious and impractical manual labelling for summarizing ultrasound videos to enhance its utility as a triage tool in the emergency department (ED) and for use in telemedicine.