计算机视觉每日论文速递[08.19]

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40篇

[检测分类相关]:

【1】 Needles in Haystacks: On Classifying Tiny Objects in Large Images
干草堆中的针:大图像中微小物体的分类
作者: Nick Pawlowski, Michal Drozdzal
链接:https://arxiv.org/abs/1908.06037

【2】 Efficient and Accurate Arbitrary-Shaped Text Detection with Pixel Aggregation Network
基于像素聚合网络的高效准确的任意形状文本检测
作者: Wenhai Wang, Chunhua Shen
备注:Accept by ICCV 2019
链接:https://arxiv.org/abs/1908.05900

【3】 GODS: Generalized One-class Discriminative Subspaces for Anomaly Detection
GUGES:用于异常检测的广义一类判别子空间
作者: Jue Wang, Anoop Cherian
备注:Accepted by ICCV 2019, 8 pages
链接:https://arxiv.org/abs/1908.05884

【4】 TASED-Net: Temporally-Aggregating Spatial Encoder-Decoder Network for Video Saliency Detection
TASED-NET:用于视频显著性检测的时间聚合空间编解码器网络
作者: Kyle Min, Jason J. Corso
备注:ICCV 2019 camera ready (Supplementary material: on CVF soon)
链接:https://arxiv.org/abs/1908.05786

【5】 Skin Lesion Segmentation and Classification for ISIC 2018 by Combining Deep CNN and Handcrafted Features
结合深度CNN和手工特征的ISIC 2018皮肤病变分割和分类
作者: Redha Ali, Temesguen Messay Kebede
链接:https://arxiv.org/abs/1908.05730

【6】 Automatic detection and diagnosis of sacroiliitis in CT scans as incidental findings
CT扫描中作为偶然发现的骶髂关节炎的自动检测和诊断
作者: Yigal Shenkman, Iris Eshed
链接:https://arxiv.org/abs/1908.05663

[分割/语义相关]:

【1】 See Clearer at Night: Towards Robust Nighttime Semantic Segmentation through Day-Night Image Conversion
参见Clear at Night:通过日夜图像转换实现稳健的夜间语义分割
作者: Lei Sun, Kaite Xiang
备注:13 pages, 7 figures, 2 tables, 2 equations. Artificial Intelligence and Machine Learning in Defense Applications, SPIE Security + Defence 2019, Strasbourg, France, September 2019
链接:https://arxiv.org/abs/1908.05868

【2】 Discretely-constrained deep network for weakly supervised segmentation
用于弱监督分割的离散约束深层网络
作者: Jizong Peng, Christian Desrosiers
链接:https://arxiv.org/abs/1908.05770

【3】 Semi-Supervised Semantic Segmentation with High- and Low-level Consistency
具有高、低级一致性的半监督语义切分
作者: Sudhanshu Mittal, Thomas Brox
链接:https://arxiv.org/abs/1908.05724

【4】 Multi-step Cascaded Networks for Brain Tumor Segmentation
多步级联网络在脑肿瘤分割中的应用
作者: Xiangyu Li, Kuanquan Wang
备注:Paper for BraTS 2019 runs in conjunction with the MICCAI 2019 conference
链接:https://arxiv.org/abs/1908.05887

[GAN/对抗式/生成式相关]:

【1】 Adversarial point perturbations on 3D objects
3D物体上的对抗性点扰动
作者: Daniel Liu, Hao Su
链接:https://arxiv.org/abs/1908.06062

【2】 FSGAN: Subject Agnostic Face Swapping and Reenactment
FSGAN:主体不可知性面孔交换和重现
作者: Yuval Nirkin, Tal Hassner
备注:2019 IEEE/CVF International Conference on Computer Vision (ICCV)
链接:https://arxiv.org/abs/1908.05932

【3】 The Angel is in the Priors: Improving GAN based Image and Sequence Inpainting with Better Noise and Structural Priors
天使在先锋:用更好的噪声和结构先锋改进基于GaN的图像和序列修复
作者: Avisek Lahiri, Prabir Kumar Biswas
链接:https://arxiv.org/abs/1908.05861

【4】 Structured Coupled Generative Adversarial Networks for Unsupervised Monocular Depth Estimation
无监督单目深度估计的结构化耦合生成对抗网络
作者: Mihai Marian Puscas, Nicu Sebe
备注:Accepted at 3DV 2019 as ORAL
链接:https://arxiv.org/abs/1908.05794

【5】 Multi-Domain Adaptation in Brain MRI through Paired Consistency and Adversarial Learning
通过成对一致性和对抗性学习实现脑MRI中的多域适应
作者: Mauricio Orbes-Arteaga, M. Jorge Cardos
备注:DART MICCAI whorshop 2019
链接:https://arxiv.org/abs/1908.05959

【6】 Incorporating human and learned domain knowledge into training deep neural networks: A differentiable dose volume histogram and adversarial inspired framework for generating Pareto optimal dose distributions in radiation therapy
将人类和学习的领域知识结合到训练深度神经网络中:可微分的剂量体积直方图和对抗性启发的框架,用于生成放射治疗中的Pareto最优剂量分布
作者: Dan Nguyen, Steve Jiang
链接:https://arxiv.org/abs/1908.05874

[行为/时空/光流/姿态/运动]:

【1】 Context-Aware Emotion Recognition Networks
情境感知情感识别网络
作者: Jiyoung Lee, Kwanghoon Sohn
备注:International Conference on Computer Vision (ICCV) 2019
链接:https://arxiv.org/abs/1908.05913

【2】 Cross-Domain Adaptation for Animal Pose Estimation
用于动物姿态估计的跨域自适应
作者: Jinkun Cao, Yu-Wing Tai
备注:accepted by ICCV'2019 for oral presentation
链接:https://arxiv.org/abs/1908.05806

【3】 DeepHuMS: Deep Human Motion Signature for 3D Skeletal Sequences
DeepHuMS:3D骨骼序列的深层人体运动特征
作者: Neeraj Battan, Avinash Sharma
链接:https://arxiv.org/abs/1908.05750

【4】 Bypass Enhancement RGB Stream Model for Pedestrian Action Recognition of Autonomous Vehicles
用于自主车辆行人行为识别的旁路增强RGB流模型
作者: Dong Cao, Lisha Xu
备注:Submitted to ACPR 2019 - Workshop on Computer Vision for Modern Vehicles
链接:https://arxiv.org/abs/1908.05674

[半/弱/无监督相关]:

【1】 Progressive Cross-camera Soft-label Learning for Semi-supervised Person Re-identification
用于半监督人再识别的渐进式交叉摄像机软标学习
作者: Lei Qi, Yang Gao
备注:arXiv admin note: text overlap with arXiv:1908.00862
链接:https://arxiv.org/abs/1908.05669

[迁移学习/domain/主动学习/自适应]:

【1】 Anchor Tasks: Inexpensive, Shared, and Aligned Tasks for Domain Adaptation
锚定任务:用于域适配的廉价、共享和对齐的任务
作者: Zhizhong Li, Derek Hoiem
链接:https://arxiv.org/abs/1908.06079

【2】 Pseudo-task Regularization for ConvNet Transfer Learning
ConvNet迁移学习的伪任务正则化
作者: Yang Zhong, Atsuto Maki
链接:https://arxiv.org/abs/1908.05997

[Re-id相关]:

【1】 Recover and Identify: A Generative Dual Model for Cross-Resolution Person Re-Identification
恢复和识别:交叉分辨人员重新识别的生成性对偶模型
作者: Yu-Jhe Li, Yu-Chiang Frank Wang
备注:Accepted to ICCV 2019
链接:https://arxiv.org/abs/1908.06052

【2】 Learning Deep Representations by Mutual Information for Person Re-identification
通过相互信息学习深度表征以进行人的重新识别
作者: Peng Chen, Dongyue Chen
链接:https://arxiv.org/abs/1908.05860

【3】 Mixed High-Order Attention Network for Person Re-Identification
用于人再识别的混合高阶注意网络
作者: Binghui Chen, Jiani Hu
备注:ICCV 2019
链接:https://arxiv.org/abs/1908.05819

[NAS相关]:

【1】 ScarletNAS: Bridging the Gap Between Scalability and Fairness in Neural Architecture Search
ScarletNAS:在神经架构搜索中弥合可伸缩性和公平性之间的鸿沟
作者: Xiangxiang Chu, Ruijun Xu
链接:https://arxiv.org/abs/1908.06022

[其他]:

【1】 Unicoder-VL: A Universal Encoder for Vision and Language by Cross-modal Pre-training
Unicoder-VL:跨模态预训练的视觉和语言通用编码器
作者: Gen Li, Ming Zhou
链接:https://arxiv.org/abs/1908.06066

【2】 Robust Principal Component Analysis for Background Estimation of Particle Image Velocimetry Data
粒子图像测速数据背景估计的稳健主成分分析
作者: Ahmadreza Baghaie
备注:Presented in LISAT 2019
链接:https://arxiv.org/abs/1908.06047

【3】 Occlusion-shared and Feature-separated Network for Occlusion Relationship Reasoning
遮挡共享特征分离网络在遮挡关系推理中的应用
作者: Rui Lu, Yu Zhou
备注:Accepted by ICCV 2019. Code and pretrained model are available at this https URL
链接:https://arxiv.org/abs/1908.05898

【4】 Zero-Shot Crowd Behavior Recognition
零炮人群行为识别
作者: Xun Xu, Timothy Hospedales
备注:Group and Crowd Behavior for Computer Vision 2017, Pages 341-369
链接:https://arxiv.org/abs/1908.05877

【5】 Differentiable Learning-to-Group Channels viaGroupable Convolutional Neural Networks
通过可分组卷积神经网络的可区分的学习到组的通道
作者: Zhaoyang Zhang, Ping Luo
备注:accepted by ICCV 2019
链接:https://arxiv.org/abs/1908.05867

【6】 daBNN: A Super Fast Inference Framework for Binary Neural Networks on ARM devices
daBNN:ARM设备上二元神经网络的超快推理框架
作者: Jianhao Zhang, Tao Mei
备注:Accepted by 2019 ACMMM Open Source Software Competition. Source code: this https URL
链接:https://arxiv.org/abs/1908.05858

【7】 Tag2Pix: Line Art Colorization Using Text Tag With SECat and Changing Loss
Tag2Pix:使用带有SECAT和更改损失的文本标签进行线条艺术着色
作者: Hyunsu Kim, Sungjoo Yoo
备注:Accepted to ICCV 2019
链接:https://arxiv.org/abs/1908.05840

【8】 Transferable Contrastive Network for Generalized Zero-Shot Learning
广义零激发学习的可转移对比网络
作者: Huajie Jiang, Xilin Chen
备注:To appear in ICCV 2019
链接:https://arxiv.org/abs/1908.05832

【9】 A Cooperative Autoencoder for Population-BasedRegularization of CNN Image Registration
一种基于群体的协同自动编码器CNN图像配准规则
作者: Riddhish Bhalodia, Ross T. Whitaker
备注:To appear in MICCAI 2019
链接:https://arxiv.org/abs/1908.05825

【10】 Empirical Bayesian Mixture Models for Medical Image Translation
医学图像翻译的经验贝叶斯混合模型
作者: Mikael Brudfors, Yael Balbastre
备注:Accepted to the Simulation and Synthesis in Medical Imaging (SASHIMI) workshop at MICCAI 2019
链接:https://arxiv.org/abs/1908.05926

【11】 Recurrent U-net: Deep learning to predict daily summertime ozone in the United States
经常性U-NET:深入学习预测美国每日夏季臭氧
作者: Tai-Long He, John R. Worden
链接:https://arxiv.org/abs/1908.05841

【12】 MimickNet, Matching Clinical Post-Processing Under Realistic Black-Box Constraints
MimickNet,现实黑盒约束下的匹配临床后处理
作者: Ouwen Huang, Mark L. Palmeri
备注:This work has been submitted to the IEEE Transactions on Medical Imaging on July 1st, 2019 for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible
链接:https://arxiv.org/abs/1908.05782

【13】 Are Quantitative Features of Lung Nodules Reproducible at Different CT Acquisition and Reconstruction Parameters?
肺结节的定量特征可以在不同的CT采集和重建参数下重现吗?
作者: Barbaros S. Erdal, Richard D. White
链接:https://arxiv.org/abs/1908.05667

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[08.19])