计算机视觉论文-2021-07-16

本专栏是计算机视觉方向论文收集积累，时间：2021年7月16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: Applying The Case Difference Heuristic to Learn Adaptations from Deep Network Features
AUTHORS: Xiaomeng Ye ; Ziwei Zhao ; David Leake ; Xizi Wang ; David Crandall
CATEGORY: cs.AI [cs.AI, cs.CV]
HIGHLIGHT: This paper investigates a two-phase process combining deep learning for feature extraction and neural network based adaptation learning from extracted features.

2, TITLE: Object Retrieval and Localization in Large Art Collections Using Deep Multi-Style Feature Fusion and Iterative Voting
AUTHORS: Nikolai Ufer ; Sabine Lang ; Bj�rn Ommer
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In the following, we introduce an algorithm that allows users to search for image regions containing specific motifs or objects and find similar regions in an extensive dataset, helping art historians to analyze large digitized art collections.

3, TITLE: Surgical Instruction Generation with Transformers
AUTHORS: Jinglu Zhang ; Yinyu Nie ; Jian Chang ; Jian Jun Zhang
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: Inspired by the neural machine translation and imaging captioning tasks in open domain, we introduce a transformer-backboned encoder-decoder network with self-critical reinforcement learning to generate instructions from surgical images.

4, TITLE: An Efficient and Small Convolutional Neural Network for Pest Recognition -- ExquisiteNet
AUTHORS: Shi-Yao Zhou ; Chung-Yen Su
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this paper, we propose a small and efficient model called ExquisiteNet to complete the task of recognizing the pests and we expect to apply our model on mobile devices.

5, TITLE: Semantic Image Cropping
AUTHORS: Oriol Corcoll
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this thesis, I introduce an additional dimension to the problem of cropping, semantics. To support my argument, I provide a new dataset containing 100 images with at least two different entities per image and four ground truth croppings collected using Amazon Mechanical Turk.

6, TITLE: Level Generation and Style Enhancement -- Deep Learning for Game Development Overview
AUTHORS: Piotr Migda? ; Bart?omiej Olechno ; B?a?ej Podg�rski
CATEGORY: cs.CV [cs.CV, I.2.10; I.4.3; J.5]
HIGHLIGHT: We present practical approaches of using deep learning to create and enhance level maps and textures for video games -- desktop, mobile, and web.

7, TITLE: What and When to Look?: Temporal Span Proposal Network for Video Visual Relation Detection
AUTHORS: Sangmin Woo ; Junhyug Noh ; Kangil Kim
CATEGORY: cs.CV [cs.CV, cs.AI]
HIGHLIGHT: To date, two representative methods have been proposed to tackle Video Visual Relation Detection (VidVRD): segment-based and window-based.

8, TITLE: Training for Temporal Sparsity in Deep Neural Networks, Application in Video Processing
AUTHORS: Amirreza Yousefzadeh ; Manolis Sifalakis
CATEGORY: cs.CV [cs.CV, cs.LG, eess.IV]
HIGHLIGHT: Towards this goal, in this paper we introduce a new DNN layer (called Delta Activation Layer), whose sole purpose is to promote temporal sparsity of activations during training.

9, TITLE: DynaDog+T: A Parametric Animal Model for Synthetic Canine Image Generation
AUTHORS: Jake Deane ; Sinead Kearney ; Kwang In Kim ; Darren Cosker
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: Consequently, we introduce a parametric canine model, DynaDog+T, for generating synthetic canine images and data which we use for a common computer vision task, binary segmentation, which would otherwise be difficult due to the lack of available data.

10, TITLE: Unsupervised Anomaly Instance Segmentation for Baggage Threat Recognition
AUTHORS: Taimur Hassan ; Samet Akcay ; Mohammed Bennamoun ; Salman Khan ; Naoufel Werghi
CATEGORY: cs.CV [cs.CV, eess.IV]
HIGHLIGHT: This paper presents a novel unsupervised anomaly instance segmentation framework that recognizes baggage threats, in X-ray scans, as anomalies without requiring any ground truth labels.

11, TITLE: Deep Learning Based Food Instance Segmentation Using Synthetic Data
AUTHORS: D. Park ; J. Lee ; J. Lee ; K. Lee
CATEGORY: cs.CV [cs.CV, cs.AI]
HIGHLIGHT: In order to solve the difficulties of data collection and annotations, this paper proposes a food segmentation method applicable to real-world through synthetic data. To perform food segmentation on healthcare robot systems, such as meal assistance robot arm, we generate synthetic data using the open-source 3D graphics software Blender placing multiple objects on meal plate and train Mask R-CNN for instance segmentation.

12, TITLE: Incorporating Lambertian Priors Into Surface Normals Measurement
AUTHORS: YAKUN JU et. al.
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this paper, we use the initial normal under the Lambertian assumption as the prior information to refine the normal measurement, instead of solely applying the observed shading cues to deriving the surface normal.

13, TITLE: MeNToS: Tracklets Association with A Space-Time Memory Network
AUTHORS: Mehdi Miah ; Guillaume-Alexandre Bilodeau ; Nicolas Saunier
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: We propose a method for multi-object tracking and segmentation (MOTS) that does not require fine-tuning or per benchmark hyperparameter selection.

14, TITLE: Lidar Light Scattering Augmentation (LISA): Physics-based Simulation of Adverse Weather Conditions for 3D Object Detection
AUTHORS: VELAT KILIC et. al.
CATEGORY: cs.CV [cs.CV, physics.optics]
HIGHLIGHT: To address this issue, we propose a physics-based approach to simulate lidar point clouds of scenes in adverse weather conditions.

15, TITLE: Potential UAV Landing Sites Detection Through Digital Elevation Models Analysis
AUTHORS: Efstratios Kakaletsis ; Nikos Nikolaidis
CATEGORY: cs.CV [cs.CV, cs.GR, 68W40]
HIGHLIGHT: In this paper, a simple technique for Unmanned Aerial Vehicles (UAVs) potential landing site detection using terrain information through identification of flat areas, is presented.

16, TITLE: A Generalized Framework for Edge-preserving and Structure-preserving Image Smoothing
AUTHORS: WEI LIU et. al.
CATEGORY: cs.CV [cs.CV, cs.GR, cs.LG]
HIGHLIGHT: A Generalized Framework for Edge-preserving and Structure-preserving Image Smoothing

17, TITLE: Recurrent Parameter Generators
AUTHORS: Jiayun Wang ; Yubei Chen ; Stella X. Yu ; Brian Cheung ; Yann LeCun
CATEGORY: cs.CV [cs.CV, cs.LG]
HIGHLIGHT: We present a generic method for recurrently using the same parameters for many different convolution layers to build a deep network.

18, TITLE: StyleFusion: A Generative Model for Disentangling Spatial Segments
AUTHORS: Omer Kafri ; Or Patashnik ; Yuval Alaluf ; Daniel Cohen-Or
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: We present StyleFusion, a new mapping architecture for StyleGAN, which takes as input a number of latent codes and fuses them into a single style code.

19, TITLE: Neighbor-view Enhanced Model for Vision and Language Navigation
AUTHORS: DONG AN et. al.
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this work, we propose a multi-module Neighbor-View Enhanced Model (NvEM) to adaptively incorporate visual contexts from neighbor views for better textual-visual matching.

20, TITLE: From Show to Tell: A Survey on Image Captioning
AUTHORS: MATTEO STEFANINI et. al.
CATEGORY: cs.CV [cs.CV, cs.CL]
HIGHLIGHT: The final goal of this work is to serve as a tool for understanding the existing state-of-the-art and highlighting the future directions for an area of research where Computer Vision and Natural Language Processing can find an optimal synergy.

21, TITLE: Training Compact CNNs for Image Classification Using Dynamic-coded Filter Fusion
AUTHORS: MINGBAO LIN et. al.
CATEGORY: cs.CV [cs.CV, cs.AI]
HIGHLIGHT: In this paper, we present a novel filter pruning method, dubbed dynamic-coded filter fusion (DCFF), to derive compact CNNs in a computation-economical and regularization-free manner for efficient image classification.

22, TITLE: STAR: Sparse Transformer-based Action Recognition
AUTHORS: FENG SHI et. al.
CATEGORY: cs.CV [cs.CV, cs.AI]
HIGHLIGHT: This work proposes a novel skeleton-based human action recognition model with sparse attention on the spatial dimension and segmented linear attention on the temporal dimension of data.

23, TITLE: Mutually Improved Endoscopic Image Synthesis and Landmark Detection in Unpaired Image-to-image Translation
AUTHORS: LALITH SHARAN et. al.
CATEGORY: cs.CV [cs.CV, cs.AI]
HIGHLIGHT: Instead, we propose to use landmark detection on the points when sutures pass into the tissue.

24, TITLE: Deep Automatic Natural Image Matting
AUTHORS: Jizhizi Li ; Jing Zhang ; Dacheng Tao
CATEGORY: cs.CV [cs.CV, cs.AI]
HIGHLIGHT: In this paper, we investigate the difficulties when extending them to natural images with salient transparent/meticulous foregrounds or non-salient foregrounds.

25, TITLE: FetalNet: Multi-task Deep Learning Framework for Fetal Ultrasound Biometric Measurements
AUTHORS: SZYMON P?OTKA et. al.
CATEGORY: cs.CV [cs.CV, cs.AI, cs.LG, eess.IV]
HIGHLIGHT: In this paper, we propose an end-to-end multi-task neural network called FetalNet with an attention mechanism and stacked module for spatio-temporal fetal ultrasound scan video analysis.

26, TITLE: Single-image Full-body Human Relighting
AUTHORS: MANUEL LAGUNAS et. al.
CATEGORY: cs.CV [cs.CV, cs.GR]
HIGHLIGHT: We present a single-image data-driven method to automatically relight images with full-body humans in them.

27, TITLE: What Image Features Boost Housing Market Predictions?
AUTHORS: Zona Kostic ; Aleksandar Jevremovic
CATEGORY: cs.CV [cs.CV, cs.LG]
HIGHLIGHT: In this paper, we propose a set of techniques for the extraction of visual features for efficient numerical inclusion in modern-day predictive algorithms.

28, TITLE: High Carbon Stock Mapping at Large Scale with Optical Satellite Imagery and Spaceborne LIDAR
AUTHORS: Nico Lang ; Konrad Schindler ; Jan Dirk Wegner
CATEGORY: cs.CV [cs.CV, cs.LG]
HIGHLIGHT: Here, we propose an automated approach that aims to support conservation and sustainable land use planning decisions by mapping tropical landscapes at large scale and high spatial resolution following the High Carbon Stock (HCS) approach.

29, TITLE: Amodal Segmentation Just Like Doing A Jigsaw
AUTHORS: Xunli Zeng ; Jianqin Yin
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: Different from this method, we propose a method of amodal segmentation based on the idea of the jigsaw.

30, TITLE: COAST: COntrollable Arbitrary-Sampling NeTwork for Compressive Sensing
AUTHORS: Di You ; Jian Zhang ; Jingfen Xie ; Bin Chen ; Siwei Ma
CATEGORY: cs.CV [cs.CV, eess.IV]
HIGHLIGHT: In this paper, we propose a novel COntrollable Arbitrary-Sampling neTwork, dubbed COAST, to solve CS problems of arbitrary-sampling matrices (including unseen sampling matrices) with one single model.

31, TITLE: StyleVideoGAN: A Temporal Generative Model Using A Pretrained StyleGAN
AUTHORS: Gereon Fox ; Ayush Tewari ; Mohamed Elgharib ; Christian Theobalt
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this paper, we present a novel approach to the video synthesis problem that helps to greatly improve visual quality and drastically reduce the amount of training data and resources necessary for generating video content.

32, TITLE: Adversarial Attacks on Multi-task Visual Perception for Autonomous Driving
AUTHORS: Ibrahim Sobh ; Ahmed Hamed ; Varun Ravi Kumar ; Senthil Yogamani
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: In this work, detailed adversarial attacks are applied on a diverse multi-task visual perception deep network across distance estimation, semantic segmentation, motion detection, and object detection.

33, TITLE: Context-Conditional Adaptation for Recognizing Unseen Classes in Unseen Domains
AUTHORS: Puneet Mangla ; Shivam Chandhok ; Vineeth N Balasubramanian ; Fahad Shahbaz Khan
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: Motivated from the success of generative zero-shot approaches, we propose a feature generative framework integrated with a COntext COnditional Adaptive (COCOA) Batch-Normalization to seamlessly integrate class-level semantic and domain-specific information.

34, TITLE: Passive Attention in Artificial Neural Networks Predicts Human Visual Selectivity
AUTHORS: THOMAS A. LANGLOIS et. al.
CATEGORY: cs.CV [cs.CV]
HIGHLIGHT: This work contributes a new approach to evaluating the biological and psychological validity of leading ANNs as models of human vision: by examining their similarities and differences in terms of their visual selectivity to the information contained in images.

35, TITLE: Variational Topic Inference for Chest X-Ray Report Generation
AUTHORS: Ivona Najdenkoska ; Xiantong Zhen ; Marcel Worring ; Ling Shao
CATEGORY: cs.CV [cs.CV, cs.LG, eess.IV]
HIGHLIGHT: The topics are inferred in a conditional variational inference framework, with each topic governing the generation of a sentence in the report. Specifically, we introduce a set of topics as latent variables to guide sentence generation by aligning image and language modalities in a latent space.

36, TITLE: Diff-Net: Image Feature Difference Based High-Definition Map Change Detection
AUTHORS: Lei He ; Shengjie Jiang ; Xiaoqing Liang ; Ning Wang ; Shiyu Song
CATEGORY: cs.CV [cs.CV, cs.RO]
HIGHLIGHT: To achieve constantly updated HD maps, we present a deep neural network (DNN), Diff-Net, to detect changes in them.

37, TITLE: Recommending Best Course of Treatment Based on Similarities of Prognostic Markers\thanks{All Authors Contributed Equally
AUTHORS: Sudhanshu ; Narinder Singh Punn ; Sanjay Kumar Sonbhadra ; Sonali Agarwal
CATEGORY: cs.IR [cs.IR, cs.CV]
HIGHLIGHT: Following this context, the goal of this paper is to propose collaborative filtering based recommender system in the healthcare sector to recommend remedies based on the symptoms experienced by the patients.

38, TITLE: MultiBench: Multiscale Benchmarks for Multimodal Representation Learning
AUTHORS: PAUL PU LIANG et. al.
CATEGORY: cs.LG [cs.LG, cs.AI, cs.CL, cs.CV, cs.MM]
HIGHLIGHT: MultiBench provides an automated end-to-end machine learning pipeline that simplifies and standardizes data loading, experimental setup, and model evaluation.

39, TITLE: Learning Sparse Interaction Graphs of Partially Observed Pedestrians for Trajectory Prediction
AUTHORS: Zhe Huang ; Ruohua Li ; Kazuki Shin ; Katherine Driggs-Campbell
CATEGORY: cs.RO [cs.RO, cs.CV]
HIGHLIGHT: Thus, we propose Gumbel Social Transformer, in which an Edge Gumbel Selector samples a sparse interaction graph of partially observed pedestrians at each time step.

40, TITLE: VILENS: Visual, Inertial, Lidar, and Leg Odometry for All-Terrain Legged Robots
AUTHORS: David Wisth ; Marco Camurri ; Maurice Fallon
CATEGORY: cs.RO [cs.RO, cs.CV]
HIGHLIGHT: We present VILENS (Visual Inertial Lidar Legged Navigation System), an odometry system for legged robots based on factor graphs.

41, TITLE: FastSHAP: Real-Time Shapley Value Estimation
AUTHORS: Neil Jethani ; Mukund Sudarshan ; Ian Covert ; Su-In Lee ; Rajesh Ranganath
CATEGORY: stat.ML [stat.ML, cs.CV, cs.LG]
HIGHLIGHT: We introduce FastSHAP, a method for estimating Shapley values in a single forward pass using a learned explainer model.

42, TITLE: A Modular U-Net for Automated Segmentation of X-ray Tomography Images in Composite Materials
AUTHORS: Jo�o P C Bertoldo ; Etienne Decenci�re ; David Ryckelynck ; Henry Proudhon
CATEGORY: eess.IV [eess.IV, cs.CV, 68T07 (Primary) 68T45 (Secondary), I.4.6; I.2.10; I.5.4; J.2]
HIGHLIGHT: In this paper a modular interpretation of UNet (Modular U-Net) is proposed and trained to segment 3D tomography images of a three-phased glass fiber-reinforced Polyamide 66.

43, TITLE: Multi-Channel Auto-Encoders and A Novel Dataset for Learning Domain Invariant Representations of Histopathology Images
AUTHORS: ANDREW MOYES et. al.
CATEGORY: eess.IV [eess.IV, cs.CV, cs.LG]
HIGHLIGHT: In this work, the Multi-Channel Auto-Encoder (MCAE) model is presented as an extension to DCAE which learns from more than two domains of data.