Yangang Wang
Ph.D. candidate
Broadband Network & Digital Multimedia Lab
Department of Automation,
Tsinghua University, Beijing 100084, China

I am currently a fifth-year Ph.D. student at Tsinghua University (THU) in Beijing, under the supervision of Prof. Dai. In 2009, I received my B.E. from the school of Instrument Science and Engineering at Southeast University (SEU). My research interests involve in the area of computer vision, computer graphics and computational photography.


2012.10 ~ 2013.03, Visiting Ph.D., EPFL, Lausanne, Switzerland.
2009.09 ~ present, Ph.D., Tsinghua University, Beijing, China.
2005.09 ~ 2009.07, B.E., Southeast University, Nanjing, China.

Research Projects
A Parametric Model for Describing the Correlation Between Single Color Images and Depth Maps
IEEE Signal Processing Letters, accepted
Yangang Wang, Ruiping Wang, Qionghai Dai
[paper][code][project page]
Abstract: This letter introduces a new approach for modeling the correlation between a single color image and its depth map with a set of parameters. The proposed model treats the color image as a set of patches and describes the correlation with a kernel function in a non-linear mapping space. We also present how to estimate the model parameters from sampled color image patches as well as the corresponding depth values. The proposed approach is tested on different color images and experimental results, which are comparable to the state-of-the-art, demonstrate the power of the proposed method. Furthermore, we validate the efficiency of the proposed model by evaluating each component of the parametric model, including the filters optimization, the choice of the patches and the kernel function.
    Video-based Hand Manipulation Capture Through Composite Motion Control
ACM Transactions on Graphics (Proc. of ACM SIGGARPH 2013), 32(4): Article NO. 43, July 2013
Yangang Wang, Jianyuan Min, Jianjie Zhang, Yebin Liu, Feng Xu, Qionghai Dai, Jinxiang Chai
[paper] [slides] [video] [project page]
Abstract: This paper describes a novel motion capture method for acquiring physically realistic hand grasping and manipulation data from multiple video streams. The key idea of our approach is to introduce a composite motion control to simultaneously model hand articulation, object movement, and subtle interaction between the hand and object. We formulate video-based hand manipulation capture in an optimization framework by maximizing the consistency between the simulated motion and the observed image data. We search an optimal motion control that drives the simulation to best match the observed image data. We demonstrate the effectiveness of our approach by capturing a wide range of high-fidelity dexterous manipulation data. The system achieves superior performance in our comparison against alternative methods such as marker-based motion capture and kinematic motion tracking. We also show the power of our recovered motion controllers by adapting the captured motion data to new objects with different properties.

    Online Modeling For Realtime Facial Animation
ACM Transactions on Graphics (Proc. of ACM SIGGARPH 2013), 32(4): Article NO. 40, July 2013
Sofien Bouaziz, Yangang Wang, Mark Pauly
[paper] [video][project page]
Abstract: We present a new algorithm for realtime face tracking on commodity RGB-D sensing devices. Our method requires no user-specific training or calibration, or any other form of manual assistance, thus enabling a range of new applications in performance-based facial animation and virtual interaction at the consumer level. The key novelty of our approach is an optimization algorithm that jointly solves for a detailed 3D expression model of the user and the corresponding dynamic tracking parameters. Realtime performance and robust computations are facilitated by a novel subspace parameterization of the dynamic facial expression space. We provide a detailed evaluation that shows that our approach significantly simplifies the performance capture workflow, while achieving accurate facial tracking for realtime applications.
High-order Markov Random Fields (MRFs)
[project page]
Abstract: In spite that high-order Markov Random Fields (MRFs) are widely used in modeling priors of natural images, their practical performance is far inferior to the theoretical limit. Targeting to learn more prior knowledge from a given database, we explore the natural image statistics at different scales and build a Normalized Filter Pool (NFP) for prior learning of nature images. Our model mainly differentiates from the previous MRFs in that we construct a multi-scale MRF model by constraining the norms of filters in another space and integrate all the filtering responses in a unified framework. we formulate both learning and inference as constrained optimization problems and solve them using augmented Lagrange method. The experiments display that the normalization of filters at different scales helps to achieve fast convergence in learning stage and obtain superior performance in image restoration tasks(including denoising and inpainting).

Personal Stuff

Honors & Awards
  • Outstanding Graduate of Southeast University, 2009
  • President Fellowship of Southeast University, 2007
  • Outstanding Student Cadre, 2006
  • First-class Scholarship of Southeast University, 2005-2008


Links Vistors
Senior apprentices Supervisor Others
Locations of visitors to this page

Last updated: Sun, 09/22/2013