计算机视觉论文-2021-07-26

本专栏是计算机视觉方向论文收集积累，时间：2021年7月26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: RGB Image Classification with Quantum Convolutional Ansaetze
AUTHORS: YU JING et. al.
CATEGORY: quant-ph [quant-ph, cs.CV, cs.LG]
HIGHLIGHT: In this paper, we propose two types of quantum circuit ansaetze to simulate convolution operations on RGB images, which differ in the way how inter-channel and intra-channel information are extracted.

2, TITLE: SurfaceNet: Adversarial SVBRDF Estimation from A Single Image
AUTHORS: Giuseppe Vecchio ; Simone Palazzo ; Concetto Spampinato
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this paper we present SurfaceNet, an approach for estimating spatially-varying bidirectional reflectance distribution function (SVBRDF) material properties from a single image. We pose the problem as an image translation task and propose a novel patch-based generative adversarial network (GAN) that is able to produce high-quality, high-resolution surface reflectance maps.

3, TITLE: Label Distribution Amendment with Emotional Semantic Correlations for Facial Expression Recognition
AUTHORS: SHASHA MAO et. al.
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: Based on this, we propose a new method that amends the label distribution of each facial image by leveraging correlations among expressions in the semantic space.

4, TITLE: Unrealistic Feature Suppression for Generative Adversarial Networks
AUTHORS: Sanghun Kim ; SeungKyu Lee
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this paper we propose unrealistic feature suppression (UFS) module that keeps high-quality features and suppresses unrealistic features.

5, TITLE: MCDAL: Maximum Classifier Discrepancy for Active Learning
AUTHORS: Jae Won Cho ; Dong-Jin Kim ; Yunjae Jung ; In So Kweon
CATEGORY: cs.CV [cs.CV, cs.AI, cs.LG]
HIGHLIGHT: In this regard, we propose a novel method to leverage the classifier discrepancies for the acquisition function for active learning.

6, TITLE: Integrating Deep Learning and Augmented Reality to Enhance Situational Awareness in Firefighting Environments
AUTHORS: Manish Bhattarai
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: We present a new four-pronged approach to build firefighter's situational awareness for the first time in the literature.

7, TITLE: Unsupervised Domain Adaptation for Video Semantic Segmentation
AUTHORS: Inkyu Shin ; Kwanyong Park ; Sanghyun Woo ; In So Kweon
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this work, we present a new video extension of this task, namely Unsupervised Domain Adaptation for Video Semantic Segmentation.

8, TITLE: Human Pose Regression with Residual Log-likelihood Estimation
AUTHORS: JIEFENG LI et. al.
CATEGORY: cs.CV [cs.CV, cs.LG]
HIGHLIGHT: In this work, we explore maximum likelihood estimation (MLE) to develop an efficient and effective regression-based methods.

9, TITLE: Reservoir Computing Approach for Gray Images Segmentation
AUTHORS: Petia Koprinkova-Hristova
CATEGORY: cs.CV [cs.CV, cs.LG, eess.IV]
HIGHLIGHT: The paper proposes a novel approach for gray scale images segmentation.

10, TITLE: Adversarial Reinforced Instruction Attacker for Robust Vision-Language Navigation
AUTHORS: BINGQIAN LIN et. al.
CATEGORY: cs.CV [cs.CV, cs.AI, cs.CL]
HIGHLIGHT: In this paper, we exploit to train a more robust navigator which is capable of dynamically extracting crucial factors from the long instruction, by using an adversarial attacking paradigm.

11, TITLE: Image-to-Image Translation with Low Resolution Conditioning
AUTHORS: Mohamed Abderrahmen Abid ; Ihsen Hedhli ; Jean-Fran�ois Lalonde ; Christian Gagne
CATEGORY: cs.CV [cs.CV, eess.IV]
HIGHLIGHT: In this work, we consider the scenario where the target image has a very low resolution.

12, TITLE: Domain Adaptive Video Segmentation Via Temporal Consistency Regularization
AUTHORS: Dayan Guan ; Jiaxing Huang ; Aoran Xiao ; Shijian Lu
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: This paper presents DA-VSN, a domain adaptive video segmentation network that addresses domain gaps in videos by temporal consistency regularization (TCR) for consecutive frames of target-domain videos.

13, TITLE: Class-Incremental Domain Adaptation with Smoothing and Calibration for Surgical Report Generation
AUTHORS: Mengya Xu ; Mobarakol Islam ; Chwee Ming Lim ; Hongliang Ren
CATEGORY: cs.CV [cs.CV, eess.IV]
HIGHLIGHT: In this work, we propose class-incremental domain adaptation (CIDA) with a multi-layer transformer-based model to tackle the new classes and domain shift in the target domain to generate surgical reports during robotic surgery.

14, TITLE: Re-distributing Biased Pseudo Labels for Semi-supervised Semantic Segmentation: A Baseline Investigation
AUTHORS: Ruifei He ; Jihan Yang ; Xiaojuan Qi
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this paper, we present a simple and yet effective Distribution Alignment and Random Sampling (DARS) method to produce unbiased pseudo labels that match the true class distribution estimated from the labeled data.

15, TITLE: Pose Estimation and 3D Reconstruction of Vehicles from Stereo-Images Using A Subcategory-Aware Shape Prior
AUTHORS: Max Coenen ; Franz Rottensteiner
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: We introduce a subcategory-aware deformable vehicle model that makes use of a prediction of the vehicle type for a more appropriate regularisation of the vehicle shape.

16, TITLE: Pre-Clustering Point Clouds of Crop Fields Using Scalable Methods
AUTHORS: Henry J. Nelson ; Nikolaos Papanikolopoulos
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this paper we notice a similarity between the current state-of-the-art for this problem and a commonly used density-based clustering algorithm, Quickshift.

17, TITLE: RewriteNet: Realistic Scene Text Image Generation Via Editing Text in Real-world Image
AUTHORS: JUNYEOP LEE et. al.
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: To address this challenge, we propose a novel representational learning-based STE model, referred to as RewriteNet that employs textual information as well as visual information.

18, TITLE: Dense Supervision Propagation for Weakly Supervised Semantic Segmentation on 3D Point Clouds
AUTHORS: Jiacheng Wei ; Guosheng Lin ; Kim-Hui Yap ; Fayao Liu ; Tzu-Yi Hung
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this paper, we train a semantic point cloud segmentation network with only a small portion of points being labeled.

19, TITLE: Score-Based Point Cloud Denoising
AUTHORS: Shitong Luo ; Wei Hu
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: To denoise a noisy point cloud, we propose to increase the log-likelihood of each point from $p * n$ via gradient ascent -- iteratively updating each point's position.

20, TITLE: Power Plant Classification from Remote Imaging with Deep Learning
AUTHORS: Michael Mommert ; Linus Scheibenreif ; Jo�lle Hanna ; Damian Borth
CATEGORY: cs.CV [cs.CV, eess.IV]
HIGHLIGHT: In this work, we focus on classifying different types of power plants from Sentinel-2 imaging data.

21, TITLE: Human Pose Transfer with Disentangled Feature Consistency
AUTHORS: KUN WU et. al.
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: To alleviate the current limitations and improve the quality of the synthesized images, we propose a pose transfer network with Disentangled Feature Consistency (DFC-Net) to facilitate human pose transfer.

22, TITLE: WaveFill: A Wavelet-based Generation Network for Image Inpainting
AUTHORS: YINGCHEN YU et. al.
CATEGORY: cs.CV [cs.CV, eess.IV]
HIGHLIGHT: This paper presents WaveFill, a wavelet-based inpainting network that decomposes images into multiple frequency bands and fills the missing regions in each frequency band separately and explicitly.

23, TITLE: Standardized Max Logits: A Simple Yet Effective Approach for Identifying Unexpected Road Obstacles in Urban-Scene Segmentation
AUTHORS: Sanghun Jung ; Jungsoo Lee ; Daehoon Gwak ; Sungha Choi ; Jaegul Choo
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: To address this issue, we propose a simple yet effective approach that standardizes the max logits in order to align the different distributions and reflect the relative meanings of max logits within each predicted class.

24, TITLE: A Deep Signed Directional Distance Function for Object Shape Representation
AUTHORS: Ehsan Zobeidi ; Nikolay Atanasov
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: This paper develops a new shape model that allows synthesizing novel distance views by optimizing a continuous signed directional distance function (SDDF).

25, TITLE: Resource Efficient Mountainous Skyline Extraction Using Shallow Learning
AUTHORS: Touqeer Ahmad ; Ebrahim Emami ; Martin ?ad�k ; George Bebis
CATEGORY: cs.CV [cs.CV, cs.AI, cs.RO, I.4.6]
HIGHLIGHT: We present a novel mountainous skyline detection approach where we adapt a shallow learning approach to learn a set of filters to discriminate between edges belonging to sky-mountain boundary and others coming from different regions.

26, TITLE: Mixed SIGNals: Sign Language Production Via A Mixture of Motion Primitives
AUTHORS: Ben Saunders ; Necati Cihan Camgoz ; Richard Bowden
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: We propose splitting the SLP task into two distinct jointly-trained sub-tasks.

27, TITLE: Pruning Ternary Quantization
AUTHORS: Dan Liu ; Xi Chen ; Jie Fu ; Xue Liu
CATEGORY: cs.CV [cs.CV, cs.AI]
HIGHLIGHT: We propose pruning ternary quantization (PTQ), a simple, yet effective, symmetric ternary quantization method.

28, TITLE: Detail Preserving Residual Feature Pyramid Modules for Optical Flow
AUTHORS: Libo Long ; Jochen Lang
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: We propose a novel Residual Feature Pyramid Module (RFPM) which retains important details in the feature map without changing the overall iterative refinement design of the optical flow estimation.

29, TITLE: Learning Discriminative Representations for Multi-Label Image Recognition
AUTHORS: Mohammed Hassanin ; Ibrahim Radwan ; Salman Khan ; Murat Tahtali
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this paper, we propose a unified deep network to learn discriminative features for the multi-label task.

30, TITLE: Transporting Causal Mechanisms for Unsupervised Domain Adaptation
AUTHORS: Zhongqi Yue ; Hanwang Zhang ; Qianru Sun ; Xian-Sheng Hua
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: To this end, we propose a practical solution: Transporting Causal Mechanisms (TCM), to identify the confounder stratum and representations by using the domain-invariant disentangled causal mechanisms, which are discovered in an unsupervised fashion.

31, TITLE: Provident Vehicle Detection at Night for Advanced Driver Assistance Systems
AUTHORS: Lukas Ewecker ; Ebubekir Asan ; Lars Ohnemus ; Sascha Saralajew
CATEGORY: cs.CV [cs.CV, cs.RO]
HIGHLIGHT: Based on previous work in this subject, we present with this paper a complete system capable of solving the task to providently detect oncoming vehicles at nighttime based on their caused light artifacts.

32, TITLE: Human Pose Estimation from Sparse Inertial Measurements Through Recurrent Graph Convolution
AUTHORS: Patrik Puchert ; Timo Ropinski
CATEGORY: cs.CV [cs.CV, cs.LG]
HIGHLIGHT: We propose the adjacency adaptive graph convolutional long-short term memory network (AAGC-LSTM) for human pose estimation from sparse inertial measurements, obtained from only 6 measurement units.

33, TITLE: Exploring Deep Registration Latent Spaces
AUTHORS: TH�O ESTIENNE et. al.
CATEGORY: cs.CV [cs.CV, cs.AI, cs.LG]
HIGHLIGHT: In this study, we investigate the topic focusing on the interpretability of deep learning-based registration methods.

34, TITLE: Tackling The Overestimation of Forest Carbon with Deep Learning and Aerial Imagery
AUTHORS: GYRI REIERSEN et. al.
CATEGORY: cs.CV [cs.CV, cs.LG]
HIGHLIGHT: This proposal paper describes the first systematic comparison of forest carbon estimation from aerial imagery, satellite imagery, and ground-truth field measurements via deep learning-based algorithms for a tropical reforestation project.

35, TITLE: Unsupervised Domain Adaptive 3D Detection with Multi-Level Consistency
AUTHORS: ZHIPENG LUO et. al.
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: Our key insight is that geometric mismatch is the key factor of domain shift.

36, TITLE: Bias Loss for Mobile Neural Networks
AUTHORS: Lusine Abrahamyan ; Valentin Ziatchin ; Yiming Chen ; Nikos Deligiannis
CATEGORY: cs.CV [cs.CV, cs.AI, cs.LG]
HIGHLIGHT: This paper proposes addressing the problem raised by random predictions by reshaping the standard cross-entropy to make it biased toward data points with a limited number of unique descriptive features.

37, TITLE: Developing Efficient Transfer Learning Strategies for Robust Scene Recognition in Mobile Robotics Using Pre-trained Convolutional Neural Networks
AUTHORS: Hermann Baumgartl ; Ricardo Buettner
CATEGORY: cs.CV [cs.CV, cs.RO]
HIGHLIGHT: We present four different robust transfer learning and data augmentation strategies for robust mobile scene recognition.

38, TITLE: Multi-Modal Pedestrian Detection with Large Misalignment Based on Modal-Wise Regression and Multi-Modal IoU
AUTHORS: Napat Wanchaitanawong ; Masayuki Tanaka ; Takashi Shibata ; Masatoshi Okutomi
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this paper, we propose a multi-modal Faster-RCNN that is robust against large misalignment.

39, TITLE: SuperCaustics: Real-time, Open-source Simulation of Transparent Objects for Deep Learning Applications
AUTHORS: Mehdi Mousavi ; Rolando Estrada
CATEGORY: cs.GR [cs.GR, cs.CV]
HIGHLIGHT: To address this issue, we present SuperCaustics, a real-time, open-source simulation of transparent objects designed for deep learning applications.

40, TITLE: Improving The Generalization of Meta-learning on Unseen Domains Via Adversarial Shift
AUTHORS: Pinzhuo Tian ; Yao Gao
CATEGORY: cs.LG [cs.LG, cs.CV]
HIGHLIGHT: In this work, we address this problem by simulating tasks from the other unseen domains to improve the generalization and robustness of meta-learning method.

41, TITLE: Data-driven Deep Density Estimation
AUTHORS: Patrik Puchert ; Pedro Hermosilla ; Tobias Ritschel ; Timo Ropinski
CATEGORY: cs.LG [cs.LG, cs.CV]
HIGHLIGHT: In this paper, we introduce a learned, data-driven deep density estimation (DDE) to infer PDFs in an accurate and efficient manner, while being independent of domain dimensionality or sample size.

42, TITLE: Domain Generalization Under Conditional and Label Shifts Via Variational Bayesian Inference
AUTHORS: XIAOFENG LIU et. al.
CATEGORY: cs.LG [cs.LG, cs.AI, cs.CV]
HIGHLIGHT: In this work, we propose a domain generalization (DG) approach to learn on several labeled source domains and transfer knowledge to a target domain that is inaccessible in training.

43, TITLE: Compositional Models: Multi-Task Learning and Knowledge Transfer with Modular Networks
AUTHORS: Andrey Zhmoginov ; Dina Bashkirova ; Mark Sandler
CATEGORY: cs.LG [cs.LG, cs.CV]
HIGHLIGHT: We propose a new approach for learning modular networks based on the isometric version of ResNet with all residual blocks having the same configuration and the same number of parameters.

44, TITLE: On The Certified Robustness for Ensemble Models and Beyond
AUTHORS: ZHUOLIN YANG et. al.
CATEGORY: cs.LG [cs.LG, cs.AI, cs.CR, cs.CV]
HIGHLIGHT: In this work, we aim to analyze and provide the certified robustness for ensemble ML models, together with the sufficient and necessary conditions of robustness for different ensemble protocols.

45, TITLE: 3D Radar Velocity Maps for Uncertain Dynamic Environments
AUTHORS: Ransalu Senanayake ; Kyle Beltran Hatch ; Jason Zheng ; Mykel J. Kochenderfer
CATEGORY: cs.RO [cs.RO, cs.CV, 68T45, I.2.9]
HIGHLIGHT: This paper explores a Bayesian approach that captures our uncertainty in the map given training data.

46, TITLE: 3D Brain Reconstruction By Hierarchical Shape-Perception Network from A Single Incomplete Image
AUTHORS: BOWEN HU et. al.
CATEGORY: eess.IV [eess.IV, cs.CV]
HIGHLIGHT: In this paper, a novel hierarchical shape-perception network (HSPN) is proposed to reconstruct the 3D point clouds (PCs) of specific brains from one single incomplete image with low latency.

47, TITLE: Explainable Artificial Intelligence (XAI) in Deep Learning-based Medical Image Analysis
AUTHORS: Bas H. M. van der Velden ; Hugo J. Kuijf ; Kenneth G. A. Gilhuijs ; Max A. Viergever
CATEGORY: eess.IV [eess.IV, cs.CV]
HIGHLIGHT: This survey presents an overview of eXplainable Artificial Intelligence (XAI) used in deep learning-based medical image analysis.

48, TITLE: Dynamic Proximal Unrolling Network for Compressive Sensing Imaging
AUTHORS: Yixiao Yang ; Ran Tao ; Kaixuan Wei ; Ying Fu
CATEGORY: eess.IV [eess.IV, cs.CV]
HIGHLIGHT: In this paper, we present a dynamic proximal unrolling network (dubbed DPUNet), which can handle a variety of measurement matrices via one single model without retraining.

49, TITLE: Photon-Starved Scene Inference Using Single Photon Cameras
AUTHORS: Bhavya Goyal ; Mohit Gupta
CATEGORY: eess.IV [eess.IV, cs.CV]
HIGHLIGHT: We propose photon scale-space a collection of high-SNR images spanning a wide range of photons-per-pixel (PPP) levels (but same scene content) as guides to train inference model on low photon flux images.

50, TITLE: AD-GAN: End-to-end Unsupervised Nuclei Segmentation with Aligned Disentangling Training
AUTHORS: Kai Yao ; Kaizhu Huang ; Jie Sun ; Curran Jude
CATEGORY: eess.IV [eess.IV, cs.CV, cs.LG]
HIGHLIGHT: To address these limitations, we propose a novel end-to-end unsupervised framework called Aligned Disentangling Generative Adversarial Network (AD-GAN).

51, TITLE: Cardiac CT Segmentation Based on Distance Regularized Level Set
AUTHORS: Xinyang Wu
CATEGORY: eess.IV [eess.IV, cs.CV]
HIGHLIGHT: In order to facilitate doctors to focus on high tech tasks such as disease analysis and diagnosis, it is crucial to develop a fast and accurate segmentation method [1].

52, TITLE: SAGE: A Split-Architecture Methodology for Efficient End-to-End Autonomous Vehicle Control
AUTHORS: Arnav Malawade ; Mohanad Odema ; Sebastien Lajeunesse-DeGroot ; Mohammad Abdullah Al Faruque
CATEGORY: eess.SP [eess.SP, cs.CV]
HIGHLIGHT: To address this problem, we propose SAGE: a methodology for selectively offloading the key energy-consuming modules of DL architectures to the cloud to optimize edge energy usage while meeting real-time latency constraints.