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34篇
[检测分类相关]:
【1】 Deep-Learning-Based Aerial Image Classification for Emergency Response Applications Using Unmanned Aerial Vehicles
基于深度学习的无人机应急应用航空图像分类
作者: Christos Kyrkou, Theocharis Theocharides
备注:CVPR International Workshop on Computer Vision for UAVs (UAVision2019), 16 June 2019
链接:https://arxiv.org/abs/1906.08716
【2】 vireoJD-MM at Activity Detection in Extended Videos
VireoJD-MM在扩展视频中的活动检测
作者: Fuchen Long, Chong-Wah Ngo
链接:https://arxiv.org/abs/1906.08547
【3】 GAN-Knowledge Distillation for one-stage Object Detection
一级目标检测的GaN知识精馏算法
作者: Wei Hong, Jingke Yu
链接:https://arxiv.org/abs/1906.08467
【4】 Nested Network with Two-Stream Pyramid for Salient Object Detection in Optical Remote Sensing Images
基于双流金字塔的嵌套网络在光学遥感图像突出目标检测中的应用
作者: Chongyi Li, Sam Kwong
链接:https://arxiv.org/abs/1906.08462
【5】 Improving the robustness of ImageNet classifiers using elements of human visual cognition
利用人类视觉认知元素提高ImageNet分类器的鲁棒性
作者: A. Emin Orhan, Brenden M. Lake
链接:https://arxiv.org/abs/1906.08416
【6】 Light Field Saliency Detection with Deep Convolutional Networks
基于深卷积网络的光场显著性检测
作者: Jun Zhang, Meng Wang
链接:https://arxiv.org/abs/1906.08331
【7】 We Need No Pixels: Video Manipulation Detection Using Stream Descriptors
我们不需要像素:使用流描述符的视频操纵检测
作者: David Güera, Edward J. Delp
备注:7 pages, 6 figures, presented at the ICML 2019 Worksop on Synthetic Realities: Deep Learning for Detecting AudioVisual Fakes
链接:https://arxiv.org/abs/1906.08743
【8】 Clustering and Classification Networks
聚类和分类网络
作者: Jin-mo Choi
链接:https://arxiv.org/abs/1906.08714
【9】 2D Linear Time-Variant Controller for Human's Intention Detection for Reach-to-Grasp Trajectories in Novel Scenes
二维线性时变控制器用于新场景中到达抓取轨迹的人的意图检测
作者: Claudio Zito, Rustam Stolkin
链接:https://arxiv.org/abs/1906.08380
[分割/语义相关]:
【1】 3D Instance Segmentation via Multi-task Metric Learning
基于多任务度量学习的三维实例分割
作者: Jean Lahoud, Martin R. Oswald
链接:https://arxiv.org/abs/1906.08650
【2】 BGrowth: an efficient approach for the segmentation of vertebral compression fractures in magnetic resonance imaging
BGrowth:磁共振成像中椎体压缩骨折分割的一种有效方法
作者: Jonathan S. Ramos, Agma J. M. Traina
链接:https://arxiv.org/abs/1906.08620
【3】 PointNLM: Point Nonlocal-Means for vegetation segmentation based on middle echo point clouds
基于中回波点云的点非局部植被分割方法
作者: Jonathan Li, Cheng Wang
链接:https://arxiv.org/abs/1906.08476
【4】 A Segmentation-Oriented Inter-Class Transfer Method: Application to Retinal Vessel Segmentation
一种面向分割的类间传递方法在视网膜血管分割中的应用
作者: Chengzhi Shi, Dali Chen
链接:https://arxiv.org/abs/1906.08501
[GAN/对抗式/生成式相关]:
【1】 Adversarial Regularization for Visual Question Answering: Strengths, Shortcomings, and Side Effects
视觉问答的对抗性正规化:优点、缺点和副作用
作者: Gabriel Grand, Yonatan Belinkov
备注:In Proceedings of the 2nd Workshop on Shortcomings in Vision and Language (SiVL) at NAACL-HLT 2019
链接:https://arxiv.org/abs/1906.08430
[行为/时空/光流/姿态/运动]:
【1】 Predicting Motion of Vulnerable Road Users using High-Definition Maps and Efficient ConvNets
利用高清晰度地图和高效ConvNet预测易受攻击道路使用者的运动
作者: Fang-Chieh Chou, Nemanja Djuric
备注:Shortened version accepted at the workshop on 'Machine Learning for Intelligent Transportation Systems' at Conference on Neural Information Processing Systems (MLITS), Montreal, Canada, 2018
链接:https://arxiv.org/abs/1906.08469
[跟踪相关]:
【1】 Performance Evaluation Methodology for Long-Term Visual Object Tracking
长期视觉目标跟踪的性能评价方法
作者: Alan Lukežič, Matej Kristan
备注:Submitted to a journal on June 2018. arXiv admin note: substantial text overlap with arXiv:1804.07056
链接:https://arxiv.org/abs/1906.08675
[裁剪/量化/加速相关]:
【1】 An Improved Trade-off Between Accuracy and Complexity with Progressive Gradient Pruning
一种改进的渐进式梯度修剪算法在精度和复杂度之间的权衡
作者: Le Thanh Nguyen-Meidine, Louis-Antoine Blais-Morin
链接:https://arxiv.org/abs/1906.08746
【2】 Reversible Privacy Preservation using Multi-level Encryption and Compressive Sensing
基于多级加密和压缩感知的可逆隐私保护
作者: Mehmet Yamac, Moncef Gabbouj
备注:5 pages, submitted/accepted, EUSIPCO 2019
链接:https://arxiv.org/abs/1906.08713
[Re-id相关]:
【1】 A Strong Baseline and Batch Normalization Neck for Deep Person Re-identification
深部人再识别的强基线和批量归一化颈部
作者: Hao Luo, Jianyang Gu
备注:This is the submitted journal version of the oral paper [arXiv:1903.07071] in CVPRW'19. Code are avaliable at: this https URL
链接:https://arxiv.org/abs/1906.08332
[其他]:
【1】 Human \textit{vs} Machine Attention in Neural Networks: A Comparative Study
人工神经网络中的机器注意:一项比较研究
作者: Qiuxia Lai, Ling Shao
链接:https://arxiv.org/abs/1906.08764
【2】 Algorithmic Guarantees for Inverse Imaging with Untrained Network Priors
无训练网络先验逆成像的算法保证
作者: Gauri Jagatap, Chinmay Hegde
链接:https://arxiv.org/abs/1906.08763
【3】 Let's Take This Online: Adapting Scene Coordinate Regression Network Predictions for Online RGB-D Camera Relocalisation
让我们在线查看:为在线RGB-D相机重新定位调整场景坐标回归网络预测
作者: Tommaso Cavallari, Stuart Golodetz
链接:https://arxiv.org/abs/1906.08744
【4】 Homogeneous Vector Capsules Enable Adaptive Gradient Descent in Convolutional Neural Networks
均匀矢量胶囊支持卷积神经网络中的自适应梯度下降
作者: Adam Byerly, Tatiana Kalganova
链接:https://arxiv.org/abs/1906.08676
【5】 Learning Generalized Transformation Equivariant Representations via Autoencoding Transformations
通过自动编码变换学习广义变换等变表示
作者: Guo-Jun Qi
备注:arXiv admin note: text overlap with arXiv:1903.10863
链接:https://arxiv.org/abs/1906.08628
【6】 Companion Surface of Danger Cylinder and its Role in Solution Variation of P3P Problem
危险圆柱体的伴面及其在P3P问题解变化中的作用
作者: Bo wang, Caixia Zhang
链接:https://arxiv.org/abs/1906.08598
【7】 Pattern Spotting in Historical Documents Using Convolutional Models
基于卷积模型的历史文献模式识别
作者: Ignacio Úbeda, Laurent Heutte
链接:https://arxiv.org/abs/1906.08580
【8】 From Zero-Shot Learning to Cold-Start Recommendation
从零启动学习到冷启动推荐
作者: v Li, Zi Huang
备注:Accepted to AAAI 2019. Codes are available at this https URL
链接:https://arxiv.org/abs/1906.08511
【9】 Multiple-Identity Image Attacks Against Face-based Identity Verification
基于人脸的身份验证的多身份图像攻击
作者: Jerone T. A. Andrews, Lewis D. Griffin
链接:https://arxiv.org/abs/1906.08507
【10】 Learning the Sampling Pattern for MRI
MRI采样模式的学习
作者: Ferdia Sherry, Matthias J. Ehrhardt
链接:https://arxiv.org/abs/1906.08754
【11】 The Limited Multi-Label Projection Layer
有限多标签投影层
作者: Brandon Amos, J. Zico Kolter
链接:https://arxiv.org/abs/1906.08707
【12】 Back to Simplicity: How to Train Accurate BNNs from Scratch?
返回简单性:如何从头开始训练准确的BNN?
作者: Joseph Bethge, Christoph Meinel
备注:Supplementary Material can be found this https URL arXiv admin note: substantial text overlap with arXiv:1812.01965
链接:https://arxiv.org/abs/1906.08637
【13】 Efficient two step optimization for large embedded deformation graph based SLAM
基于SLAM的大型嵌入变形图的有效两步优化
作者: Jingwei Song, Rong Xiong
链接:https://arxiv.org/abs/1906.08477
【14】 SwiftNet: Using Graph Propagation as Meta-knowledge to Search HighlyvRepresentative Neural Architectures
SwiftNet:利用图的传播作为元知识搜索高代表性的神经结构
作者: Hsin-Pai (Dave) Cheng, Yiran Chen
链接:https://arxiv.org/abs/1906.08305
【15】 Training on test data: Removing near duplicates in Fashion-MNIST
测试数据培训:消除时尚中的近乎重复项-MNIST
作者: Christopher Geier
链接:https://arxiv.org/abs/1906.08255
翻译:腾讯翻译君