Conference Program (download pdf)

Tuesday, Apr. 11: Side Events
14:00 – 17:00 Young Scientists Forum
19:00 – 20:20 Student Training: Research and Scientific Paper Writing
Wednesday, Apr. 12: Registration & Side Events
09:00 – 20:00 Conference Registration, Lecture Hall
09:00 – 17:20 Smart Robotic Workshop
17:40 – 19:00 Reception of CVM
 Day One Thursday, Apr.13, 2017 Lecture Hall
08:30 – 09:20 Opening & Photographing
09:20 – 10:00 Invited talk I: From Quad Meshes to Quad Layouts,  by Leif Kobbelt. Chair: Shi-Min Hu
Session 1 Retrieval, Chair: Shi-Min Hu
10:00 – 10:20 Non-negative Locality-constrained Vocabulary Tree for Finger Vein Image Retrieval, Kun Su, Gongping Yang, Lu Yang, Peng Su and Yilong Yin
10:20 – 10:40 Medical signs recognition of lung nodules based on image retrieval with semantic feature and supervised hashing, Juanjuan Zhao, Ling Pan, Pengfei Zhao and Xiaoxian Tang
10:40 – 11:00 Coffee break
Session 2 Detection, Chair: Peter Hall
11:00 – 11:20 Joint Salient Object Detection and Existence Prediction, Huaizu Jiang, Ming-Ming Cheng, Shi-Jie Li, Ali Borji and Jingdong Wang
11:20 – 11:40 Crowd Counting via Learning Perspective in Multi-scale Multi-view Web Images, Chong Shang and Haizhou Ai
11:40 – 12:00 Collective Representation for Abnormal Detection, Renzhen Ye and Xuelong Li
12:00 – 12:20 Robust Tracking-by-Detection Using a Selector and Refinement Mechanism, Ruochen Fan and Fanglue Zhang
12:20 – 14:00 Lunch
Session 3 Image Enhancement, Chair: Hung-Kuo Chu
14:00 – 14:20 Automatic colorization with improved spatial coherence and boundary localization, Wei Zhang, Chaowei Fang and Guanbin Li
14:20 – 14:40 Photographic Appearance Enhancement via Detail-based Dictionary Learning, Zhifeng Xie and Lizhuang Ma
14:40 – 15:00 User-guided Line Abstraction using Coherence and Structure Analysis, Hui-Chi Tsai, Ya-Hsuan Lee, Hung-Kuo Chu and Ruen-Rone Lee
15:00 – 15:20 Multi-Example Feature-Constrained Back-Projection for Image-Resolution, Junlei Zhang, Xuemei Li and Xin Zhang
15:20 – 15:40 Coffee break
Session 4 Learning Methodology, Chair: Youfu Li
15:40 – 16:00 Graph Regularized Low-Rank Representation for Semi-Supervised Learning, Cong-Zhe You and Xiao-Jun Wu
16:00 – 16:20 Semi-supervised dictionary learning with label propagation for image classification, Lin Chen and Meng Yang
16:20 – 16:40 Discriminative Histogram Intersection Metric Learning and Its Applications, Pengyi Hao, Yang Xia, Sei-Ichiro Kamata, Xiao-Xin Li and Shengyong Chen
16:40 – 17:00 EasySVM: A Visual Analysis Approach for Open-Box Support Vector Machines, Yuxin Ma, Wei Chen, Xiaohong Ma, Jiayi Xu, Xinxin Huang, Ross Maciejewski and Anthony K. H. Tung
Session 5 Fast Presentation for Poster (2 minutes each), Chair: Dong-Ming Yan
17:00 – 17:24 1. Object-Aware Image Editing, Shiming Ge, Xin Jin, Qiting Ye and Zhao Luo
2. MSEdge: A Multi-Scale Edge Chain Detector, Xiaohu Lu, Jian Yao, Xiaofeng Zhang, Li Li and Yahui Liu
3. Batch Image Alignment via Subspace Recovery Based on Alternative Sparsity Pursuit, Xianhui Lin, Zhu Liang Yu, Zhenghui Gu and Zhaoquan Cai
4. Feature-aware variational subdivision surface reconstruction, Xiaoqun Wu, Jianmin Zheng, Yiyu Cai and Haisheng Li
5. A Procedural Texture Generation Framework Based on Semantic Descriptions, Junyu Dong, Lina Wang, Jun Liu and Xin Sun
6. Foreground Object Extraction Based on Graph and Depth Layers, Zhiguang Xiao, Hui Chen, Changhe Tu and Reinhard Klette
7. Visual Tracking via Convolutional Network and Structured Output Support Vector Machine, Junwei Li, Xiaolong Zhou, Sixian Chan and Shengyong Chen
8. Face image retrieval based on shape and texture feature fusion, Zongguang Lu, Jing Yang and Qingshan Liu
9. A hierarchical structure of co-occurrence relationship for Occlusion handling, Xiaowei Zhang and Bo Li
10. Joint Head Pose and Face Landmarks Regression from Depth Images, Jie Wang, Juyong Zhang, Changwei Luo and Falai Chen
11. ILPN: An Independent Losses Pose Net for Globally Locating Body Joints, Le Dong, Xiuyuan Chen, Ran Wang, Wenpu Dong, Bo Hu and Ebroul Izquierdo
12. Vectorial Approximations of Innite-Dimensional Covariance Descriptors for Image Classication, Jieyi Ren and Xiaojun Wu
17:24 – 18:00 Poster Discussion
18:00 – 20:00 Banquet
Day Two Friday, Apr.14, 2017 Lecture Hall
09:00 – 09:40 Invited talk II: Toward Deep Geometric Image Understanding,  by Jia Deng. Chair: Ralph Martin
Session 6 Segmentation, Chair: Ralph Martin
09:40 – 10:00 Feature-Aligned Segmentation using Correlation Clustering, Yixin Zhuang, Hang Dou, Nathan Carr and Tao Ju
10:00 – 10:20 Texture Region Segmentation from Manga, Xueting Liu, Chengze Li and Tien-Tsin Wong
10:20 – 10:40 Prior-free Dependent Motion Segmentation using Helmholtz-Hodge Decomposition based Object-Motion Oriented Map, Cuicui Zhang and Zhilei Liu
10:40 – 11:00 Coffee break
Session 7 Scene Understanding, Chair: Song-Hai Zhang
11:00 – 11:20 Static Scene Illumination Estimation from Video with Applications,Bin Liu, Ralph Martin, Kun Xu and Shi-Min Hu
11:20 – 11:40 Fast and Accurate Visual Odometry from A Monocular Camera, Xin Yang
11:40 – 12:00 Feature-based RGB-D camera pose optimization for real-time 3D reconstruction, Chao Wang and Xiaohu Guo
12:00 – 12:20 Temporally Consistent Depth Map Prediction Using Deep CNN and Spatial-temporal CRF, Xuran Zhao, Xun Wang and Qichao Chen
12:20 – 14:00 Lunch
Session 8 Modeling, Chair: Yu-Kun Lai
14:00 – 14:20 ExploreTree: Interactive Tree Modeling in Semantic Trait Space with Online Intent Learning, Yinhui Yang, Rui Wang, Hongxin Zhang and Hujun Bao
14:20 – 14:40 Minkowski Sum Computation of B-spline Surfaces, Jonathan Mizrahi, Sijoon Kim, Iddo Hanniel, Myung-Soo Kim and Gershon Elber
14:40 – 15:00 Rigidity Controllable As-Rigid-As-Possible Shape Deformation, Shu-Yu Chen, Lin Gao, Yu-Kun Lai and Shihong Xia
15:00 – 15:20 A Fast Propagation Scheme for Approximate Geodesic Paths, Xiaoguang Han, Hongchuan Yu and Jianjun Zhang
15:20 – 15:40 Coffee Break
Session 9 Video Processing, Chair: Yizhou Yu
15:40 – 16:00 Practical automatic background substitution for live video, Haozhi Huang, Xiaonan Fang, Yufei Ye, Songhai Zhang, Paul Rosin and Shimin Hu
16:00 – 16:20 Captioning Videos using Large-scale Image Corpus, Xiaoyu Du, Yang Yang, Liu Yang, Fumin Shen and Jinhui Tang
16:20 – 16:40 Robust Facial Landmark Detection and Tracking across Poses and Expressions for In-The-Wild Monocular Video, Yongqiang Zhang, Shuang Liu, Xiaosong Yang, Daming Shi and Jian Jun Zhang
16:40 – 17:00 Closing session


Keynote Speaker I:

Prof. Leif Kobbelt  

Talk Title: From Quad Meshes to Quad Layouts

Talk Abstract:

The conversion of raw geometric data (that typically comes in the form of unstructured triangle meshes) to high quality quad meshes is an important and challenging task. The complexity of the task results from the fact that quad mesh topologies are subject to global consistency requirements which cannot be dealt with by local constructions. This is why recent quad meshing techniques formulate the mesh generation process as a global optimization problem. By adding hard and soft constraints to this optimization, many desired properties such as structural simplicity, principal direction alignment, as well as injectivity can be guaranteed by construction. An even more challenging problem is the computation of quad layouts, where a coarse segmentation of the input surface into essentially rectangular patches is sought which also satisfies global consistency and shape quality requirements. While being structurally related, both problems need to be addressed by fundamentally different approaches. In my talk I will present some of these approaches and demonstrate that they can generate high quality quad meshes and quad layouts with a high degree of automation but that they also allow the user to interactively control the results by setting boundary conditions accordingly.

Short Bio:

Leif Kobbelt is a full professor of Computer Science with a specialization in Computer Graphics and Geometry Processing. Since 2001 he is the head of the Institute for Computer Graphics and Multimedia at RWTH Aachen University which in 2015 has been extended into the Visual Computing Institute.

After having received his diploma in 1992 and his PhD in 1994 in Computer Science from the Karlsruhe Institute of Technology he worked at the University of Wisconsin in Madison, the University of Erlangen-Nuremberg and the Max Planck Institute of Computer Science before he moved to RWTH Aachen University in 2001. His major research interests include 3D reconstruction, efficient geometry processing, realistic real-time rendering and (mobile) multimedia applications. Kobbelt published a substantial number of influential papers in international top-conferences and journals. In addition, he acts as a consultant, reviewer, and editor for international companies, research organizations and journals respectively. For his research he was awarded with a number of renowned academic prices including the Heinz-Maier-Leibnitz price in 2000, the Eurographics Outstanding Technical Contribution Award 2004, the Günther Enderle Award (in 1999 and 2012), an ERC Advanced Grant 2013 and the Gottfried Wilhelm Leibniz Prize in 2014. He has been named a Fellow of the Eurographics Association (2008) and a Distinguished Professor of RWTH Aachen University (2013). In 2015 he became a member of the Academia Europaea and in 2016 a member of the North Rhine Westphalian Academy of Sciences, Humanities and the Arts.


Keynote Speaker II:

Prof. Jia Deng

Talk Title:Toward Deep Geometric Image Understanding

Talk Abstract:

Achieving human-level visual understanding requires extracting richer geometric information from images. In particular, it entails moving beyond 2D bounding boxes of objects to more detailed geometric representations. In this talk I will present recent work in this direction. In particular, I will discuss human pose estimation (keypoint localization) and single-image depth estimation. For human pose estimation, I will present a new type of convolutional network architecture called the stacked hourglass network. For single-image depth estimation, I will show how to use crowdsourcing to improve depth estimation for images in the wild.

Short Bio:

Jia Deng is an Assistant Professor of Computer Science and Engineering at the University of Michigan. His research focus is on computer vision and machine learning, in particular, achieving human-level visual understanding by integrating perception, cognition, and learning. He received his Ph.D. from Princeton University and his B.Eng. from Tsinghua University, both in computer science. He is a recipient of the PAMI Everingham Prize, the Yahoo ACE Award, a Google Faculty Research Award, the ICCV Marr Prize, and the ECCV Best Paper Award.