同步wx订阅号(arXiv每日论文速递),支持后台回复'search 关键词'查询相关的最新论文。有些许帮助的话,麻烦关注一下哦(* ̄rǒ ̄)
cs.CV 方向,今日共计45篇
[检测分类相关]:
【1】 How much real data do we actually need: Analyzing object detection performance using synthetic and real data
我们实际需要多少真实数据:使用合成数据和真实数据分析对象检测性能
作者: Farzan Erlik Nowruzi, Julien Rebut
备注:Accepted in International Conference on Machine Learning (ICML 2019) Workshop on AI for Autonomous Driving
链接:https://arxiv.org/abs/1907.07061
【2】 Fused Detection of Retinal Biomarkers in OCT Volumes
OCT体积中视网膜生物标志物的融合检测
作者: Thomas Kurmann, Raphael Sznitman
链接:https://arxiv.org/abs/1907.06955
【3】 Semi-supervised Breast Lesion Detection in Ultrasound Video Based on Temporal Coherence
基于时间相干的超声视频半监督乳腺病变检测
作者: Sihong Chen, Yefeng Zheng
链接:https://arxiv.org/abs/1907.06941
【4】 Mango Tree Net -- A fully convolutional network for semantic segmentation and individual crown detection of mango trees
芒果树网-一个用于芒果树语义分割和个体树冠检测的完全卷积网络
作者: Vikas Agaradahalli Gurumurthy, Omkar Narasipura
链接:https://arxiv.org/abs/1907.06915
【5】 Cascade RetinaNet: Maintaining Consistency for Single-Stage Object Detection
级联RetinaNet:保持单级对象检测的一致性
作者: Hongkai Zhang, Xilin Chen
备注:BMVC 2019
链接:https://arxiv.org/abs/1907.06881
【6】 Rethinking RGB-D Salient Object Detection: Models, Datasets, and Large-Scale Benchmarks
重新思考RGB-D凸起对象检测:模型、数据集和大规模基准测试
作者: Deng-Ping Fan, Ming-Ming Cheng
链接:https://arxiv.org/abs/1907.06781
【7】 Improving 3D Object Detection for Pedestrians with Virtual Multi-View Synthesis Orientation Estimation
利用虚拟多视点综合方位估计改进行人三维目标检测
作者: Jason Ku, Steven L. Waslander
备注:Accepted in IROS 2019
链接:https://arxiv.org/abs/1907.06777
【8】 AugLabel: Exploiting Word Representations to Augment Labels for Face Attribute Classification
AugLabel:利用单词表示来增加用于人脸属性分类的标签
作者: Binod Bhattarai, Tae-Kyun Kim
链接:https://arxiv.org/abs/1907.06757
【9】 Explaining Classifiers with Causal Concept Effect (CaCE)
用因果概念效应解释量词
作者: Yash Goyal, Been Kim
链接:https://arxiv.org/abs/1907.07165
[分割/语义相关]:
【1】 Efficient Segmentation: Learning Downsampling Near Semantic Boundaries
有效的分割:在语义边界附近学习下采样
作者: Dmitrii Marin, Yuri Boykov
链接:https://arxiv.org/abs/1907.07156
【2】 Uncertainty-aware Self-ensembling Model for Semi-supervised 3D Left Atrium Segmentation
半监督三维左房分割的不确定性感知自集成模型
作者: Lequan Yu, Pheng-Ann Heng
备注:Accepted by MICCAI2019; Code is available in this https URL
链接:https://arxiv.org/abs/1907.07034
【3】 Data Selection for training Semantic Segmentation CNNs with cross-dataset weak supervision
跨数据集弱监督的语义分割CNN训练数据选择
作者: Panagiotis Meletis, Gijs Dubbelman
备注:IEEE ITSC 2019
链接:https://arxiv.org/abs/1907.07023
【4】 Improving Semantic Segmentation via Dilated Affinity
通过扩展亲和度改进语义切分
作者: Boxi Wu, Deng Cai
备注:10 pages, 5 figures, under review of NIPS2019
链接:https://arxiv.org/abs/1907.07011
【5】 Separable Convolutional LSTMs for Faster Video Segmentation
用于更快视频分割的可分离卷积LSTM
作者: Andreas Pfeuffer, Klaus Dietmayer
链接:https://arxiv.org/abs/1907.06876
【6】 Stereo-based terrain traversability analysis using normal-based segmentation and superpixel surface analysis
使用基于法线的分割和超像素表面分析的基于立体的地形可穿透性分析
作者: Aras R. Dargazany
链接:https://arxiv.org/abs/1907.06823
【7】 Real-time Hair Segmentation and Recoloring on Mobile GPUs
移动GPU上的实时头发分割和重新着色
作者: Andrei Tkachenka, Siargey Pisarchyk
备注:4 pages, 5 figures; CVPR Workshop on Computer Vision for Augmented and Virtual Reality, Long Beach, CA, USA, 2019
链接:https://arxiv.org/abs/1907.06740
【8】 MaskPlus: Improving Mask Generation for Instance Segmentation
MaskPlus:改进实例分割的掩码生成
作者: Shichao Xu, Qi Zhu
链接:https://arxiv.org/abs/1907.06713
【9】 Anatomically-Informed Multiple Linear Assignment Problems for White Matter Bundle Segmentation
基于解剖学的白质束分割的多重线性分配问题
作者: Giulia Bertò, Emanuele Olivetti
链接:https://arxiv.org/abs/1907.07077
【10】 CLCI-Net: Cross-Level fusion and Context Inference Networks for Lesion Segmentation of Chronic Stroke
CLCI-NET:用于慢性中风病变分割的跨级融合和上下文推理网络
作者: Hao Yang, Shanshan Wang
链接:https://arxiv.org/abs/1907.07008
【11】 X-Net: Brain Stroke Lesion Segmentation Based on Depthwise Separable Convolution and Long-range Dependencies
X-NET:基于Depthwise可分离卷积和长程依赖的脑卒中病变分割
作者: Kehan Qi, Shanshan Wang
链接:https://arxiv.org/abs/1907.07000
【12】 AirwayNet: A Voxel-Connectivity Aware Approach for Accurate Airway Segmentation Using Convolutional Neural Networks
AirwayNet:一种利用卷积神经网络进行精确气道分割的体素连通性感知方法
作者: Yulei Qin, Guang-Zhong Yang
链接:https://arxiv.org/abs/1907.06852
[GAN/对抗式/生成式相关]:
【1】 On the ''steerability" of generative adversarial networks
论生成对抗网络的“可操作性”
作者: Ali Jahanian, Phillip Isola
链接:https://arxiv.org/abs/1907.07171
【2】 Natural Adversarial Examples
自然对抗性例子
作者: Dan Hendrycks, Dawn Song
链接:https://arxiv.org/abs/1907.07174
【3】 Adversarial Sensor Attack on LiDAR-based Perception in Autonomous Driving
自主驾驶中对抗性传感器对LiDAR感知的攻击
作者: Yulong Cao, Z. Morley Mao
链接:https://arxiv.org/abs/1907.06826
[行为/时空/光流/姿态/运动]:
【1】 Speed estimation evaluation on the KITTI benchmark based on motion and monocular depth information
基于运动和单目深度信息的Kitti基准速度估计评估
作者: Róbert-Adrian Rill
链接:https://arxiv.org/abs/1907.06989
【2】 A Unified Deep Framework for Joint 3D Pose Estimation and Action Recognition from a Single RGB Camera
一种用于单个RGB相机的联合3D姿态估计和动作识别的统一深度框架
作者: Huy Hieu Pham, Sergio A Velastin
链接:https://arxiv.org/abs/1907.06968
【3】 Human Pose Estimation for Real-World Crowded Scenarios
真实拥挤场景下的人体姿态估计
作者: Thomas Golda, Jürgen Beyerer
链接:https://arxiv.org/abs/1907.06922
【4】 Instant Motion Tracking and Its Applications to Augmented Reality
即时运动跟踪及其在增强现实中的应用
作者: Jianing Wei, Tingbo Hou
备注:CVPR Workshop on Computer Vision for Augmented and Virtual Reality, Long Beach, CA, 2019
链接:https://arxiv.org/abs/1907.06796
【5】 Slow Feature Analysis for Human Action Recognition
人体动作识别的慢特征分析
作者: Zhang Zhang, Dacheng Tao
链接:https://arxiv.org/abs/1907.06670
[跟踪相关]:
【1】 Pedestrian Tracking by Probabilistic Data Association and Correspondence Embeddings
基于概率数据关联和对应嵌入的行人跟踪
作者: Borna Bićanić, Ivan Petrović
链接:https://arxiv.org/abs/1907.07045
[迁移学习/domain/主动学习相关]:
【1】 Learning Depth from Monocular Videos Using Synthetic Data: A Temporally-Consistent Domain Adaptation Approach
使用合成数据从单目视频中学习深度:一种时间一致的域适应方法
作者: Yipeng Mou, Dacheng Tao
链接:https://arxiv.org/abs/1907.06882
[裁剪/量化/加速相关]:
【1】 Single-bit-per-weight deep convolutional neural networks without batch-normalization layers for embedded systems
嵌入式系统无批归一化层的单比特深度卷积神经网络
作者: Mark D. McDonnell, Andre van Schaik
链接:https://arxiv.org/abs/1907.06916
【2】 An Inter-Layer Weight Prediction and Quantization for Deep Neural Networks based on a Smoothly Varying Weight Hypothesis
基于平滑变权假设的深层神经网络层间权值预测与量化
作者: Kang-Ho Lee, Sung-Ho Bae
链接:https://arxiv.org/abs/1907.06835
[数据集dataset]:
【1】 A Short Note on the Kinetics-700 Human Action Dataset
关于Kinetics-700人类行为数据集的简短说明
作者: Joao Carreira, Andrew Zisserman
备注:arXiv admin note: substantial text overlap with arXiv:1808.01340
链接:https://arxiv.org/abs/1907.06987
[其他视频相关]:
【1】 Real-time Facial Surface Geometry from Monocular Video on Mobile GPUs
基于移动GPU的单目视频实时人脸表面几何
作者: Yury Kartynnik, Matthias Grundmann
备注:4 pages, 4 figures; CVPR Workshop on Computer Vision for Augmented and Virtual Reality, Long Beach, CA, USA, 2019
链接:https://arxiv.org/abs/1907.06724
[其他]:
【1】 Predicting Next-Season Designs on High Fashion Runway
预测时尚T台下一季的设计
作者: Yusan Lin, Hao Yang
链接:https://arxiv.org/abs/1907.07161
【2】 EnforceNet: Monocular Camera Localization in Large Scale Indoor Sparse LiDAR Point Cloud
EnforceNet:大规模室内稀疏LiDAR点云中的单目摄像机定位
作者: Yu Chen, Guan Wang
链接:https://arxiv.org/abs/1907.07160
【3】 Perception of visual numerosity in humans and machines
人类和机器对视觉数量的感知
作者: Alberto Testolin, Marco Zorzi
链接:https://arxiv.org/abs/1907.06996
【4】 A General Framework for Uncertainty Estimation in Deep Learning
深度学习中不确定性估计的一般框架
作者: Mattia Segù, Davide Scaramuzza
链接:https://arxiv.org/abs/1907.06890
【5】 Deep inspection: an electrical distribution pole parts study via deep neural networks
深度检测:基于深度神经网络的配电杆件研究
作者: Liangchen Liu, Brian Lovell
备注:electrical distribution pole inspection, integrated inspection system, object detection, imbalanced data classification, To appear in Proceeding of ICIP 2019
链接:https://arxiv.org/abs/1907.06844
【6】 2nd Place Solution to the GQA Challenge 2019
2019 GQA挑战赛第二名解决方案
作者: Shijie Geng, Dimitris N. Metaxas
链接:https://arxiv.org/abs/1907.06794
【7】 Efficient Pipeline for Camera Trap Image Review
用于相机陷印图像审查的高效流水线
作者: Sara Beery, Siyu Yang
备注:From the Data Mining and AI for Conservation Workshop at KDD19
链接:https://arxiv.org/abs/1907.06772
【8】 Boosting Resolution and Recovering Texture of micro-CT Images with Deep Learning
利用深度学习提高显微CT图像的分辨率和纹理恢复
作者: Ying Da Wang, Peyman Mostaghimi
链接:https://arxiv.org/abs/1907.07131
【9】 Improved Reinforcement Learning through Imitation Learning Pretraining Towards Image-based Autonomous Driving
基于图像自主驾驶的模拟学习预训练改进强化学习
作者: Tianqi Wang, Dong Eui Chang
备注:5 pages, 2019 19th International Conference on Control, Automation and Systems (ICCAS 2019)
链接:https://arxiv.org/abs/1907.06838
【10】 Deep learning-based color holographic microscopy
基于深度学习的彩色全息显微术
作者: Tairan Liu, Aydogan Ozcan
链接:https://arxiv.org/abs/1907.06727
翻译:腾讯翻译君