计算机视觉每日论文速递[08.21]

同步公众号(arXiv每日论文速递),回复'search 关键词'查询相关最新论文。(* ̄rǒ ̄)

cs.CV 方向,今日共计42篇

[检测分类相关]:

【1】 Towards High-Resolution Salient Object Detection
面向高分辨率凸起目标检测
作者: Yi Zeng, Huchuan Lu
备注:Accepted by ICCV2019. The HRSOD dataset is available at this https URL
链接:https://arxiv.org/abs/1908.07274

【2】 Human uncertainty makes classification more robust
人类的不确定性使分类更加健壮
作者: Joshua C. Peterson, Olga Russakovsky
备注:In Proceedings of the 2019 IEEE International Conference on Computer Vision (ICCV)
链接:https://arxiv.org/abs/1908.07086

【3】 Endotracheal Tube Detection and Segmentation in Chest Radiographs using Synthetic Data
使用合成数据的胸片中气管内管的检测和分割
作者: Maayan Frid-Adar, Hayit Greenspan
备注:Accepted to MICCAI 2019
链接:https://arxiv.org/abs/1908.07170

【4】 Learning Fixed Points in Generative Adversarial Networks: From Image-to-Image Translation to Disease Detection and Localization
生成对抗性网络中的学习不动点:从图像到图像的翻译到疾病检测和定位
作者: Md Mahfuzur Rahman Siddiquee, Jianming Liang
链接:https://arxiv.org/abs/1908.06965

[分割/语义相关]:

【1】 Learning Semantic-Specific Graph Representation for Multi-Label Image Recognition
用于多标签图像识别的学习语义特定图表示
作者: Tianshui Chen, Liang Lin
备注:accepted by ICCV 2019
链接:https://arxiv.org/abs/1908.07325

【2】 Unsupervised Multi-modal Style Transfer for Cardiac MR Segmentation
用于心脏MR分割的无监督多模态风格转换
作者: Chen Chen, Daniel Rueckert
备注:Accepted at STACOM 2019 held in conjunction with MICCAI 2019
链接:https://arxiv.org/abs/1908.07344

[GAN/对抗式/生成式相关]:

【1】 RelGAN: Multi-Domain Image-to-Image Translation via Relative Attributes
RelGAN:通过相对属性实现多域图像到图像的转换
作者: Po-Wei Wu, Shih-Wei Liao
备注:Accepted to ICCV 2019
链接:https://arxiv.org/abs/1908.07269

【2】 A Neural Virtual Anchor Synthesizer based on Seq2Seq and GAN Models
基于Seq2Seq和GaN模型的神经虚拟锚点合成器
作者: Zipeng Wang, Shiguo Lian
链接:https://arxiv.org/abs/1908.07262

【3】 Boundless: Generative Adversarial Networks for Image Extension
无穷:用于图像扩展的生成性对抗性网络
作者: Piotr Teterwak, William T. Freeman
链接:https://arxiv.org/abs/1908.07007

[行为/时空/光流/姿态/运动]:

【1】 Skeleton-based Gait Index Estimation with LSTMs
基于骨架的LSTM步态指数估计
作者: Trong Nguyen Nguyen, Jean Meunier
备注:2018 IEEE/ACIS 17th International Conference on Computer and Information Science (ICIS)
链接:https://arxiv.org/abs/1908.07416

【2】 Estimating skeleton-based gait abnormality index by sparse deep auto-encoder
利用稀疏深度自动编码器估计基于骨架的步态异常指数
作者: Trong Nguyen Nguyen, Jean Meunier
备注:2018 IEEE Seventh International Conference on Communications and Electronics (ICCE)
链接:https://arxiv.org/abs/1908.07415

【3】 ViSiL: Fine-grained Spatio-Temporal Video Similarity Learning
ViSiL:细粒度时空视频相似性学习
作者: Giorgos Kordopatis-Zilos, Ioannis Kompatsiaris
链接:https://arxiv.org/abs/1908.07410

【4】 Human Mesh Recovery from Monocular Images via a Skeleton-disentangled Representation
通过骨架解缠表示从单目图像中恢复人体网格
作者: Sun Yu, Mei Tao
备注:Accepted by ICCV2019
链接:https://arxiv.org/abs/1908.07172

【5】 Resolving 3D Human Pose Ambiguities with 3D Scene Constraints
利用3D场景约束解决3D人体姿态模糊性
作者: Mohamed Hassan, Michael J. Black
备注:To appear in ICCV 2019
链接:https://arxiv.org/abs/1908.06963

【6】 Autonomous, Monocular, Vision-Based Snake Robot Navigation and Traversal of Cluttered Environments using Rectilinear Gait Motion
自主,单目,基于视觉的蛇形机器人使用直线步态运动导航和穿越杂乱的环境
作者: Alexander H. Chang (1), Patricio A. Vela (1) ((1) Georgia Institute of Technology)
链接:https://arxiv.org/abs/1908.07101

[跟踪相关]:

【1】 An End-to-end Video Text Detector with Online Tracking
一种具有在线跟踪功能的端到端视频文本检测器
作者: Hongyuan Yu, Liang Wang
链接:https://arxiv.org/abs/1908.07135

[迁移学习/domain/主动学习/自适应]:

【1】 Customizing Student Networks From Heterogeneous Teachers via Adaptive Knowledge Amalgamation
通过自适应知识融合从异构教师定制学生网络
作者: Chengchao Shen, Mingli Song
备注:Accepted to ICCV 2019
链接:https://arxiv.org/abs/1908.07121

[超分辨率]:

【1】 SROBB: Targeted Perceptual Loss for Single Image Super-Resolution
SROBB:针对单幅图像超分辨率的目标感知损失
作者: Mohammad Saeed Rad, Jean-Philippe Thiran
备注:ICCV 2019
链接:https://arxiv.org/abs/1908.07222

[点云]:

【1】 BoxNet: A Deep Learning Method for 2D Bounding Box Estimation from Bird's-Eye View Point Cloud
Boxnet:一种基于鸟眼视点云的2D包围盒估计深度学习方法
作者: Ehsan Nezhadarya, Bingbing Liu
链接:https://arxiv.org/abs/1908.07085

[深度depth相关]:

【1】 Human Gait Symmetry Assessment using a Depth Camera and Mirrors
利用深度相机和反射镜进行人体步态对称性评估
作者: Trong-Nguyen Nguyen, Jean Meunier
链接:https://arxiv.org/abs/1908.07422

【2】 Assessment of gait normality using a depth camera and mirrors
使用深度照相机和镜子评估步态正常
作者: Trong Nguyen Nguyen, Jean Meunier
备注:2018 IEEE EMBS International Conference on Biomedical & Health Informatics (BHI)
链接:https://arxiv.org/abs/1908.07418

[人脸相关]:

【1】 Make a Face: Towards Arbitrary High Fidelity Face Manipulation
做一张脸:走向任意高保真度的脸部操作
作者: Shengju Qian, Ran He
备注:Accepted to ICCV 2019
链接:https://arxiv.org/abs/1908.07191

[3D/3D重建等相关]:

【1】 Probabilistic Reconstruction Networks for 3D Shape Inference from a Single Image
基于单幅图像的三维形状推断的概率重建网络
作者: Roman Klokov, Edmond Boyer
备注:Accepted as Oral at BMVC'19
链接:https://arxiv.org/abs/1908.07475

【2】 Joint Embedding of 3D Scan and CAD Objects
3D扫描与CAD对象的联合嵌入
作者: Manuel Dahnert, Matthias Nießner
链接:https://arxiv.org/abs/1908.06989

[其他视频相关]:

【1】 Proposal-free Temporal Moment Localization of a Natural-Language Query in Video using Guided Attention
使用引导注意的视频中自然语言查询的无建议时间矩定位
作者: Cristian Rodriguez Opazo, Stephen Gould
链接:https://arxiv.org/abs/1908.07236

[其他]:

【1】 Image Synthesis From Reconfigurable Layout and Style
基于可重构布局和风格的图像合成
作者: Wei Sun, Tianfu Wu
备注:Accepted to ICCV 2019
链接:https://arxiv.org/abs/1908.07500

【2】 Pix2Pose: Pixel-Wise Coordinate Regression of Objects for 6D Pose Estimation
Pix2Pose:用于6D姿态估计的对象的像素-Wise坐标回归
作者: Kiru Park, Markus Vincze
备注:Accepted at ICCV 2019 (Oral)
链接:https://arxiv.org/abs/1908.07433

【3】 Learning to Sit: Synthesizing Human-Chair Interactions via Hierarchical Control
学会坐下:通过分级控制合成人-椅互动
作者: Yu-Wei Chao, Jia Deng
链接:https://arxiv.org/abs/1908.07423

【4】 Blind Image Deconvolution using Pretrained Generative Priors
利用预训练生成先验知识的盲图象解卷积
作者: Muhammad Asim, Ali Ahmed
备注:Accepted in BMVC 2019. Extended version of this paper can be found at arXiv:1802.04073
链接:https://arxiv.org/abs/1908.07404

【5】 Consistent Scale Normalization for Object Recognition
目标识别的一致尺度归一化
作者: Zewen He, Wensheng Zhang
链接:https://arxiv.org/abs/1908.07323

【6】 Three-D Safari: Learning to Estimate Zebra Pose, Shape, and Texture from Images "In the Wild"
三维野生动物园:学习从“野外”图像估计斑马姿势,形状和纹理
作者: Silvia Zuffi, Michael J. Black
备注:Accepted at ICCV 2019
链接:https://arxiv.org/abs/1908.07201

【7】 Zero-Shot Grounding of Objects from Natural Language Queries
自然语言查询中对象的零点扎根
作者: Arka Sadhu, Ram Nevatia
备注:ICCV19 oral, camera-ready version
链接:https://arxiv.org/abs/1908.07129

【8】 360-Degree Textures of People in Clothing from a Single Image
从单个图像中获取服装中人的360度纹理
作者: Verica Lazova, Gerard Pons-Moll
链接:https://arxiv.org/abs/1908.07117

【9】 Unpaired Image-to-Speech Synthesis with Multimodal Information Bottleneck
具有多模态信息瓶颈的非配对图像语音合成
作者: Shuang Ma, Yale Song
备注:ICCV 2019
链接:https://arxiv.org/abs/1908.07094

【10】 UprightNet: Geometry-Aware Camera Orientation Estimation from Single Images
UprightNet:基于单幅图像的几何感知相机方位估计
作者: Wenqi Xian, Noah Snavely
链接:https://arxiv.org/abs/1908.07070

【11】 LXMERT: Learning Cross-Modality Encoder Representations from Transformers
LXMERT:从变压器中学习跨模态编码器表示
作者: Hao Tan, Mohit Bansal
备注:EMNLP 2019 (12 pages)
链接:https://arxiv.org/abs/1908.07490

【12】 Playing magic tricks to deep neural networks untangles human deception
玩魔术让深层神经网络解开人类的欺骗
作者: Regina Zaghi-Lara, Alex Gomez-Marin
链接:https://arxiv.org/abs/1908.07446

【13】 Cross-modal Zero-shot Hashing
跨模态零射散列
作者: Xuanwu Liu, Xiangliang Zhang
链接:https://arxiv.org/abs/1908.07388

【14】 A Novel method for IDC Prediction in Breast Cancer Histopathology images using Deep Residual Neural Networks
一种利用深度残差神经网络预测乳腺癌组织病理学图像IDC的新方法
作者: Chandra Churh Chatterjee, Gopal Krishna
备注:Accepted at 2nd International Conference on Intelligent Communication and Computational Techniques,2019
链接:https://arxiv.org/abs/1908.07362

【15】 n-MeRCI: A new Metric to Evaluate the Correlation Between Predictive Uncertainty and True Error
n-merci:一种评估预测不确定度与真实误差相关性的新度量
作者: Michel Moukari, Frédéric Jurie
链接:https://arxiv.org/abs/1908.07253

【16】 StateLens: A Reverse Engineering Solution for Making Existing Dynamic Touchscreens Accessible
StateLens:使现有动态触摸屏可访问的逆向工程解决方案
作者: Anhong Guo, Jeffrey P. Bigham
备注:ACM UIST 2019
链接:https://arxiv.org/abs/1908.07144

【17】 Protecting Neural Networks with Hierarchical Random Switching: Towards Better Robustness-Accuracy Trade-off for Stochastic Defenses
用分层随机切换保护神经网络:走向更好的鲁棒性-随机防御的精确度权衡
作者: Xiao Wang, Peter Chin
备注:Published as Conference Paper @ IJCAI 2019
链接:https://arxiv.org/abs/1908.07116

翻译:腾讯翻译君

你可能感兴趣的:(计算机视觉每日论文速递[08.21])