计算机视觉/图像处理论文速递[02.24]

格林先生MrGreen arXiv每日学术速递

计算机视觉/图像处理每日论文速递[02.24]

[检测分类相关]:

【1】 Robust Iris Presentation Attack Detection Fusing 2D and 3D Information
融合2D和3D信息的鲁棒虹膜表示攻击检测
作者:Zhaoyuan Fang,  Kevin W. Bowyer
链接:arxiv.org/abs/2002.09137

【2】 Detection and Classification of Astronomical Targets with Deep Neural  Networks in Wide Field Small Aperture Telescopes
大视场小孔径望远镜天文目标的深度神经网络检测与分类
作者:Peng Jia,  Yongyang Sun
链接:arxiv.org/abs/2002.09211

【3】 Comparing Different Deep Learning Architectures for Classification of  Chest Radiographs
不同深度学习架构对胸部X线片分类的比较
作者:Keno K. Bressem,  Janis Vahldiek
链接:arxiv.org/abs/2002.08991

[分割/语义相关]:

【1】 3D U-Net for Segmentation of Plant Root MRI Images in Super-Resolution
三维U-net超分辨率植物根系MRI图像分割
作者:Yi Zhao,  Sven Behnke
链接:arxiv.org/abs/2002.09317

【2】 SemanticPOSS: A Point Cloud Dataset with Large Quantity of Dynamic  Instances
语义POSS:一个具有大量动态实例的点云数据集
作者:Yancheng Pan,  Huijing Zhao
备注:submited to IEEE Intelligent Vehicles Symposium(2020)
链接:arxiv.org/abs/2002.09147 [人脸相关]:

【1】 Deep Multi-Facial Patches Aggregation Network For Facial Expression  Recognition
面向面部表情识别的深度多面片聚集网络
作者:Ahmed Rachid Hazourli,  Alice Othmani
备注:arXiv admin note: substantial text overlap with arXiv:1909.10305
链接:arxiv.org/abs/2002.09298

【2】 Unsupervised Enhancement of Soft-biometric Privacy with Negative Face  Recognition
基于阴性人脸识别的软生物特征隐私的无监督增强
作者:Philipp Terhörst,  Arjan Kuijper
链接:arxiv.org/abs/2002.09181

【3】 Face Phylogeny Tree Using Basis Functions
基于基函数的人脸系统发育树
作者:Sudipta Banerjee,  Arun Ross
链接:arxiv.org/abs/2002.09068

【4】 Cortical surface parcellation based on intra-subject white matter fiber  clustering
基于主体内白质纤维聚类的大脑皮层表面分块
作者:Narciso López-López,  Pamela Guevara
备注:This research has received funding from the European Union's Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie Actions H2020-MSCA-RISE-2015 BIRDS GA No. 690941, CONICYT PFCHA/ DOCTORADO NACIONAL/2016-21160342, CONICYT FONDECYT 1190701, CONICYT PIA/Anillo de Investigaci'on en Ciencia y Tecnolog'ia ACT172121 and CONICYT Basal Center FB0008
链接:arxiv.org/abs/2002.09034

[GAN/对抗式/生成式相关]:

【1】 Domain Adaptive Adversarial Learning Based on Physics Model Feedback for  Underwater Image Enhancement
基于物理模型反馈的领域自适应对抗学习水下图像增强
作者:Yuan Zhou,  Kangming Yan
链接:arxiv.org/abs/2002.09315

【2】 Cross-Resolution Adversarial Dual Network for Person Re-Identification  and Beyond
交叉分辨对抗性双重网络用于人的重新识别和超越
作者:Yu-Jhe Li,  Yu-Chiang Frank Wang
备注:Submitted to IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI). 16 pages. arXiv admin note: substantial text overlap with arXiv:1908.06052
链接:arxiv.org/abs/2002.09274

【3】 BlockGAN: Learning 3D Object-aware Scene Representations from Unlabelled  Images
BlockGAN:从未标记的图像中学习3D对象感知场景表示
作者:Thu Nguyen-Phuoc,  Niloy Mitra
链接:arxiv.org/abs/2002.08988

【4】 Bidirectional Generative Modeling Using Adversarial Gradient Estimation
使用对抗性梯度估计的双向生成建模
作者:Xinwei Shen,  Kani Chen
链接:arxiv.org/abs/2002.09161

[图像/视频检索]:

【1】 Fine-Grained Instance-Level Sketch-Based Video Retrieval
基于细粒度实例级草图的视频检索
作者:Peng Xu,  Yi-Zhe Song
链接:arxiv.org/abs/2002.09461

[行为/时空/光流/姿态/运动]:

【1】 Human Action Recognition using Local Two-Stream Convolution Neural  Network Features and Support Vector Machines
基于局部双流卷积神经网络特征和支持向量机的人体动作识别
作者:David Torpey,  Turgay Celik
链接:arxiv.org/abs/2002.09423

【2】 Audio-video Emotion Recognition in the Wild using Deep Hybrid Networks
基于深度混合网络的野外音视频情感识别
作者:Xin Guo,  Kenneth E. Barner
链接:arxiv.org/abs/2002.09023

【3】 Convolutional Tensor-Train LSTM for Spatio-temporal Learning
用于时空学习的卷积张量列LSTM
作者:Jiahao Su,  Animashree Anandkumar
链接:arxiv.org/abs/2002.09131

[半/弱/无监督相关]:

【1】 Unsupervised Pre-trained, Texture Aware And Lightweight Model for Deep  Learning-Based Iris Recognition Under Limited Annotated Data
有限标注数据下基于深度学习的虹膜识别的无监督预训练、纹理感知和轻量级模型
作者:Manashi Chakraborty,  Pabitra Mitra
备注:Under review at ICIP2020
链接:arxiv.org/abs/2002.09048

[裁剪/量化/加速相关]:

【1】 Post-training Quantization with Multiple Points: Mixed Precision without  Mixed Precision
训练后多点量化:不带混合精度的混合精度
作者:Xingchao Liu,  Qiang Liu
链接:arxiv.org/abs/2002.09049

[数据集dataset]:

【1】 The DIDI dataset: Digital Ink Diagram data
滴滴数据集:数字墨水图数据
作者:Philippe Gervais,  Otmar Hilliges
链接:arxiv.org/abs/2002.09303

[3D/3D重建等相关]:

【1】 Leveraging Photogrammetric Mesh Models for Aerial-Ground Feature Point  Matching Toward Integrated 3D Reconstruction
利用摄影测量网格模型进行航空-地面特征点匹配以实现集成三维重建
作者:Qing Zhu,  Yeting Zhang
链接:arxiv.org/abs/2002.09085

【2】 Learning Precise 3D Manipulation from Multiple Uncalibrated Cameras
从多个未校准的摄像机学习精确的3D操作
作者:Iretiayo Akinola,  Dmitry Kalashnikov
备注:Accepted at International Conference on Robotics and Automation (ICRA 2020)
链接:arxiv.org/abs/2002.09107

[其他视频相关]:

【1】 SummaryNet: A Multi-Stage Deep Learning Model for Automatic Video  Summarisation
SummaryNet:一种面向视频自动摘要的多阶段深度学习模型
作者:Ziyad Jappie,  Turgay Celik
链接:arxiv.org/abs/2002.09424

【2】 Stochastic Latent Residual Video Prediction
随机潜在残差视频预测
作者:Jean-Yves Franceschi (MLIA),  Patrick Gallinari (MLIA)
链接:arxiv.org/abs/2002.09219

【3】 Disentangling Controllable Object through Video Prediction Improves  Visual Reinforcement Learning
通过视频预测解缠可控对象改进了视觉强化学习
作者:Yuanyi Zhong,  Jian Peng
备注:Accepted to ICASSP 2020
链接:arxiv.org/abs/2002.09136 [其他]:

【1】 The Automated Inspection of Opaque Liquid Vaccines
不透明液体疫苗的自动检测
作者:Gregory Palmer,  Harry Flore
备注:8 pages, 5 Figures, 3 Tables, ECAI 2020 Conference Proceedings
链接:arxiv.org/abs/2002.09406

【2】 A Convolutional Neural Network into graph space
图空间中的卷积神经网络
作者:Maxime Martineau,  Gilles Venturini
链接:arxiv.org/abs/2002.09285

【3】 Learning to Inpaint by Progressively Growing the Mask Regions
通过逐步增加蒙版区域来学习绘画
作者:Mohamed Abbas Hedjazi,  Yakup Genc
备注:ICCV Workshop on Should we preregister experiments in computer vision?, Seoul, South Korea, 2019
链接:arxiv.org/abs/2002.09280

【4】 Affective Expression Analysis in-the-wild using Multi-Task Temporal  Statistical Deep Learning Model
基于多任务时态统计深度学习模型的情感表达分析
作者:Nhu-Tai Do,  Soo-Hyung Kim
链接:arxiv.org/abs/2002.09120

【5】 Adapted Center and Scale Prediction: More Stable and More Accurate
自适应中心和尺度预测:更稳定和更准确
作者:Wenhao Wang
链接:arxiv.org/abs/2002.09053

【6】 Brain Age Estimation Using LSTM on Children's Brain MRI
LSTM在儿童脑MRI上的脑年龄估计
作者:Sheng He,  Yangming Ou
备注:ISBI 2020
链接:arxiv.org/abs/2002.09045

【7】 Complete Endomorphisms in Computer Vision
计算机视觉中的完全自同态
作者:Javier Finat,  Francisco Delgado-del-Hoyo
链接:arxiv.org/abs/2002.09003

【8】 Are Gabor Kernels Optimal for Iris Recognition?
Gabor内核对于虹膜识别是最佳的吗?
作者:Aidan Boyd,  Kevin Bowyer
备注:To appear at IJCB 2020
链接:arxiv.org/abs/2002.08959

【9】 Fast Implementation of Morphological Filtering Using ARM NEON Extension
利用ARM霓虹灯扩展快速实现形态滤波
作者:Elena Limonova,  Vladimir Arlazarov
链接:arxiv.org/abs/2002.09474

【10】 Calibrating Deep Neural Networks using Focal Loss
利用焦损校准深神经网络
作者:Jishnu Mukhoti,  Puneet K. Dokania
链接:arxiv.org/abs/2002.09437

【11】 Efficient Learning of Model Weights via Changing Features During  Training
通过在训练过程中改变特征有效地学习模型权重
作者:Marcell Beregi-Kovács,  András Hajdu
链接:arxiv.org/abs/2002.09249

【12】 Exploiting the Full Capacity of Deep Neural Networks while Avoiding  Overfitting by Targeted Sparsity Regularization
充分利用深度神经网络的容量,同时避免目标稀疏度正则化的过拟合
作者:Karim Huesmann,  Benjamin Risse
链接:arxiv.org/abs/2002.09237

【13】 Curating Social Media Data
策划社交媒体数据
作者:Kushal Vaghani
链接:arxiv.org/abs/2002.09202

【14】 Residual Knowledge Distillation
剩余知识蒸馏
作者:Mengya Gao,  Chen Change Loy
链接:arxiv.org/abs/2002.09168

【15】 Greedy Policy Search: A Simple Baseline for Learnable Test-Time  Augmentation
贪婪策略搜索:可学习测试时间增强的简单基线
作者:Dmitry Molchanov,  Dmitry Vetrov
链接:arxiv.org/abs/2002.09103

【16】 Comparing recurrent and convolutional neural networks for predicting  wave propagation
比较递归神经网络和卷积神经网络预测波的传播
作者:Stathi Fotiadis,  Anil A. Bharath
链接:arxiv.org/abs/2002.08981

【17】 Learning Intermediate Features of Object Affordances with a  Convolutional Neural Network
用卷积神经网络学习目标函数的中间特征
作者:Aria Yuan Wang,  Michael J. Tarr
备注:Published on 2018 Conference on Cognitive Computational Neuroscience. See <this https URL>
链接:arxiv.org/abs/2002.08975

【18】 Affinity and Diversity: Quantifying Mechanisms of Data Augmentation
亲和性与多样性:数据增强的量化机制
作者:Raphael Gontijo-Lopes,  Ethan Dyer
链接:arxiv.org/abs/2002.08973

机器翻译,仅供参考


文章推荐