【今日CV 视觉论文速览】05 Dec 2018

今日CS.CV计算机视觉论文速览

Wed, 5 Dec 2018

Totally 58 papers

Interesting:

• 基于单张图片预测完整点云 （from 阿姆斯特丹大学CV Lab）

  

  模型如下所示：

  

  首先通过2D RGB图生成出深度图，其作用在于规范2D-3D域之间的转换，并限制学习到的流型结构。同时基于深度图生成部分点云图,从像素坐标转到空间坐标

  Z [ u , v , 1 ] T = K ( R [ X , Y , Z ] T + t ) Z[u,v,1]^T = K(R[X,Y,Z]^T+t) Z[u,v,1]T=K(R[X,Y,Z]T+t)

  随后，通过在部分点云空间和完整点云空间的转化来获取点云。

  在抽取部分点云特征矢量的基础上，利用类似PCN的方法实现了稀疏但完整点云的输出。（基于全连接，负责几何外形预测）

  随后基于folding解码器来生成稠密点云（负责近似表面和局域形貌）

  最后利用预测出的完整点云来约束部分点云的预测误差。

  模型预测结果：

  

  实际图像结果：

  

• AutoFocus: Efficient Multi-Scale Inference,提出了一种高效的多尺度目标检测算法用于高效检测小物体。这种算法使用了由粗到精的策略，只在那些可能有小物体存在的区域使用细粒度的检测。为了得到这些区域，本文提出了一种称为FocusPixels的方法来预测小区域。同时为了配合FocusPixels高效的使用，研究人员给出了FocusChip来涵盖FocusPixels区域，以减少计算量。(from 马里兰大学)

  

  上图所示，不同尺度的下，只有较小的大象被标记为FocusPixels

  

  图像的处理流程如上图所示。

  

  SNIPER
• 从单张图像恢复点云,主要将这一重建过程分解为图像到深度图、深度度通过相机模型到部分点云，再到稠密点云。
• 基于深度Inception生成网络的认知图像修复,这篇文章提出了方法克服了先前方法单一感受野的限制和人工痕迹的生成。多感受野可以改善图像特征的抽象能力，池化可以保持特征不变性。inception模块用于提取高层信息，并且利用了不同的mask数据集来提高修复模型的适应性。

包含inception模块的生成网络 （from 同济）：

结果如下：

结果如图所示：

• 基于改进的WGAN实现语义图像修复 （from University Pompeu Fabra）

  

  得到结果如下：

  

Daily Computer Vision Papers

[1] Title: Learning 3D Human Dynamics from Video

Authors:Angjoo Kanazawa, Jason Zhang, Panna Felsen, Jitendra Malik

[2] Title: AutoFocus: Efficient Multi-Scale Inference

Authors:Mahyar Najibi, Bharat Singh, Larry S. Davis

[3] Title: Monocular Total Capture: Posing Face, Body, and Hands in the Wild

Authors:Donglai Xiang, Hanbyul Joo, Yaser Sheikh

[4] Title: Improving Semantic Segmentation via Video Propagation and Label Relaxation

Authors:Yi Zhu, Karan Sapra, Fitsum A. Reda, Kevin J. Shih, Shawn Newsam, Andrew Tao, Bryan Catanzaro

[5] Title: Detect-to-Retrieve: Efficient Regional Aggregation for Image Search

Authors:Marvin Teichmann, Andre Araujo, Menglong Zhu, Jack Sim

[6] Title: A novel database of Children’s Spontaneous Facial Expressions (LIRIS-CSE)

Authors:Rizwan Ahmed Khan, Crenn Arthur, Alexandre Meyer, Saida Bouakaz

[7] Title: Towards generative adversarial networks as a new paradigm for radiology education

Authors:Samuel G. Finlayson, Hyunkwang Lee, Isaac S. Kohane, Luke Oakden-Rayner

[8] Title: A Face-to-Face Neural Conversation Model

Authors:Hang Chu, Daiqing Li, Sanja Fidler

[9] Title: SurfConv: Bridging 3D and 2D Convolution for RGBD Images

Authors:Hang Chu, Wei-Chiu Ma, Kaustav Kundu, Raquel Urtasun, Sanja Fidler

[10] Title: Content Authentication for Neural Imaging Pipelines: End-to-end Optimization of Photo Provenance in Complex Distribution Channels

Authors:Pawel Korus, Nasir Memon

[11] Title: PolyMapper: Extracting City Maps using Polygons

Authors:Zuoyue Li, Jan Dirk Wegner, Aurélien Lucchi

[12] Title: Sturm: Sparse Tubal-Regularized Multilinear Regression for fMRI

Authors:Wenwen Li, Jian Lou, Shuo Zhou, Haiping Lu

[13] Title: Cross-spectral Periocular Recognition: A Survey

Authors:S. S. Behera, Bappaditya Mandal, N. B. Puhan

[14] Title: Feasibility of Colon Cancer Detection in Confocal Laser Microscopy Images Using Convolution Neural Networks

Authors:Nils Gessert, Lukas Wittig, Daniel Drömann, Tobias Keck, Alexander Schlaefer, David B. Ellebrecht

[15] Title: The Visual Centrifuge: Model-Free Layered Video Representations

Authors:Jean-Baptiste Alayrac, João Carreira, Andrew Zisserman

[16] Title: Deep Inception Generative Network for Cognitive Image Inpainting

Authors:Qingguo Xiao, Guangyao Li, Qiaochuan Chen

[17] Title: Inferring Point Clouds from Single Monocular Images by Depth Intermediation

Authors:Wei Zeng, Sezer Karaoglu, Theo Gevers

[18] Title: Meta Learning Deep Visual Words for Fast Video Object Segmentation

Authors:Harkirat Singh Behl, Mohammad Najafi, Philip H.S. Torr

[19] Title: TextField: Learning A Deep Direction Field for Irregular Scene Text Detection

Authors:Yongchao Xu, Yukang Wang, Wei Zhou, Yongpan Wang, Zhibo Yang, Xiang Bai

[20] Title: A Deep Learning Framework for Semi-Supervised Cross-Modal Retrieval with Label Prediction

Authors:Devraj Mandal, Pramod Rao, Soma Biswas

[21] Title: Estimating 6D Pose From Localizing Designated Surface Keypoints

Authors:Zelin Zhao, Gao Peng, Haoyu Wang, Hao-Shu Fang, Chengkun Li, Cewu Lu

[22] Title: Weakly Supervised Convolutional LSTM Approach for Tool Tracking in Laparoscopic Videos

Authors:Chinedu Innocent Nwoye, Didier Mutter, Jacques Marescaux, Nicolas Padoy

[23] Title: From biological vision to unsupervised hierarchical sparse coding

Authors:Victor Boutin, Angelo Franciosini, Franck Ruffier, Laurent. U Perrinet

[24] Title: Timeception for Complex Action Recognition

Authors:Noureldien Hussein, Efstratios Gavves, Arnold W.M. Smeulders

[25] Title: FaceFeat-GAN: a Two-Stage Approach for Identity-Preserving Face Synthesis

Authors:Yujun Shen, Bolei Zhou, Ping Luo, Xiaoou Tang

[26] Title: Rare Event Detection using Disentangled Representation Learning

Authors:Ryuhei Hamaguchi, Ken Sakurada, Ryosuke Nakamura

[27] Title: Towards Continuous Domain adaptation for Healthcare

Authors:Rahul Venkataramani, Hariharan Ravishankar, Saihareesh Anamandra

[28] Title: Learning to Explain with Complemental Examples

Authors:Atsushi Kanehira, Tatsuya Harada

[29] Title: Image Dehazing via Joint Estimation of Transmittance Map and Environmental Illumination

Authors:Sanchayan Santra, Ranjan Mondal, Pranoy Panda, Nishant Mohanty, Shubham Bhuyan

[30] Title: Multimodal Explanations by Predicting Counterfactuality in Videos

Authors:Atsushi Kanehira, Kentaro Takemoto, Sho Inayoshi, Tatsuya Harada

[31] Title: Conditional Video Generation Using Action-Appearance Captions

Authors:Shohei Yamamoto, Antonio Tejero-de-Pablos, Yoshitaka Ushiku, Tatsuya Harada

[32] Title: Factorized Attention: Self-Attention with Linear Complexities

Authors:Zhuoran Shen, Mingyuan Zhang, Shuai Yi, Junjie Yan, Haiyu Zhao

[33] Title: Classifying Collisions with Spatio-Temporal Action Graph Networks

Authors:Roei Herzig, Elad Levi, Huijuan Xu, Eli Brosh, Amir Globerson, Trevor Darrell

[34] Title: The Alignment of the Spheres: Globally-Optimal Spherical Mixture Alignment for Camera Pose Estimation

Authors:Dylan Campbell, Lars Petersson, Laurent Kneip, Hongdong Li, Stephen Gould

[35] Title: Ladder Networks for Semi-Supervised Hyperspectral Image Classification

Authors:Julian Büchel, Okan Ersoy

[36] Title: Zoom-In-to-Check: Boosting Video Interpolation via Instance-level Discrimination

Authors:Liangzhe Yuan, Yibo Chen, Hantian Liu, Tao Kong, Jianbo Shi

[37] Title: Walking on Thin Air: Environment-Free Physics-based Markerless Motion Capture

Authors:Micha Livne, Leonid Sigal, Marcus A. Brubaker, David J. Fleet

[38] Title: Learning to Fuse Things and Stuff

Authors:Jie Li, Allan Raventos, Arjun Bhargava, Takaaki Tagawa, Adrien Gaidon

[39] Title: Bag of Tricks for Image Classification with Convolutional Neural Networks

Authors:Junyuan Xie, Tong He, Zhi Zhang, Hang Zhang, Zhongyue Zhang, Mu Li

[40] Title: Deep Generative Modeling of LiDAR Data

Authors:Lucas Caccia, Herke van Hoof, Aaron Courville, Joelle Pineau

[41] Title: Cross-Classification Clustering: An Efficient Multi-Object Tracking Technique for 3-D Instance Segmentation in Connectomics

Authors:Yaron Meirovitch, Lu Mi, Hayk Saribekyan, Alexander Matveev, David Rolnick, Casimir Wierzynski, Nir Shavit

[42] Title: Disease Detection in Weakly Annotated Volumetric Medical Images using a Convolutional LSTM Network

Authors:Nathaniel Braman, David Beymer, Ehsan Dehghan

[43] Title: ZerNet: Convolutional Neural Networks on Arbitrary Surfaces via Zernike Local Tangent Space Estimation

Authors:Zhiyu Sun, Jia Lu, Stephen Baek

[44] Title: Crowd Sourcing based Active Learning Approach for Parking Sign Recognition

Authors:Humayun Irshad, Qazaleh Mirsharif, Jennifer Prendki

[45] Title: Semantic Image Inpainting Through Improved Wasserstein Generative Adversarial Networks

Authors:Patricia Vitoria, Joan Sintes, Coloma Ballester

[46] Title: Machine Friendly Machine Learning: Interpretation of Computed Tomography Without Image Reconstruction

Authors:Hyunkwang Lee, Chao Huang, Sehyo Yune, Shahein H. Tajmir, Myeongchan Kim, Synho Do

[47] Title: QR code denoising using parallel Hopfield networks

Authors:Ishan Bhatnagar, Shubhang Bhatnagar

[48] Title: MS-ASL: A Large-Scale Data Set and Benchmark for Understanding American Sign Language

Authors:Hamid Reza Vaezi Joze, Oscar Koller

[49] Title: Brain Tumor Segmentation using an Ensemble of 3D U-Nets and Overall Survival Prediction using Radiomic Features

Authors:Xue Feng, Nicholas Tustison, Craig Meyer

[50] Title: Identification and Recognition of Rice Diseases and Pests Using Deep Convolutional Neural Networks

Authors:Chowdhury Rafeed Rahman, Preetom Saha Arko, Mohammed Eunus Ali, Mohammad Ashik Iqbal Khan, Abu Wasif, Md. Rafsan Jani, Md. Shahjahan Kabir

[51] Title: A Two-Stream Variational Adversarial Network for Video Generation

Authors:Ximeng Sun, Huijuan Xu, Kate Saenko

[52] Title: DeepVoxels: Learning Persistent 3D Feature Embeddings

Authors:Vincent Sitzmann, Justus Thies, Felix Heide, Matthias Nießner, Gordon Wetzstein, Michael Zollhöfer

[53] Title: Disentangling Latent Hands for Image Synthesis and Pose Estimation

Authors:Linlin Yang, Angela Yao

[54] Title: Compressive Classification (Machine Learning without learning)

Authors:Vincent Schellekens, Laurent Jacques

[55] Title: A multi-class structured dictionary learning method using discriminant atom selection

Authors:R.E. Rolón, L.E. Di Persia, R.D. Spies, H.L. Rufiner

[56] Title: Prototype-based Neural Network Layers: Incorporating Vector Quantization

Authors:Sascha Saralajew, Lars Holdijk, Maike Rees, Thomas Villmann

[57] Title: Improving Traffic Safety Through Video Analysis in Jakarta, Indonesia

Authors:João Caldeira, Alex Fout, Aniket Kesari, Raesetje Sefala, Joseph Walsh, Katy Dupre, Muhammad Rizal Khaefi, Setiaji, George Hodge, Zakiya Aryana Pramestri, Muhammad Adib Imtiyazi

[58] Title: A Hybrid Instance-based Transfer Learning Method

Authors:Azin Asgarian, Parinaz Sobhani, Ji Chao Zhang, Madalin Mihailescu, Ariel Sibilia, Ahmed Bilal Ashraf, Babak Taati

Papers from arxiv.org

pic from pixels.com