YU Zhiding

 

M.Phil. Candidate

Multimedia Technology Research Center (MTrec)
Dept. of Electronic and Computer Engineering, The Hong Kong Univ. of Science and Technology
Room 6518, Academic Building, HKUST, Clear Water Bay, Kowloon, HK

Research Associate

Human Sensing Laboratory

The Robotics Institute, School of Computer Science, Carnegie Mellon University
EDSH 110, Carnegie Mellon University, 5000 Forbes Ave, Pittsburgh, PA, 15213

 

Email: yzhiding AT andrew.cmu.edu; zdyu AT ust.hk

I will  apply for the 2012 fall intake of PhD in Electrical Engineering / Computer Science. The Curriculum Vitae is available here: [PDF]

Biography

I received the B.Eng. degree from the School of Electronic & Information Engineering (Talented Student Program), South China University of Technology (SCUT) in 2008, and took a one-year M.Sc. study at the Department of Electronic & Computer Engineering, The Hong Kong University of Science and Technology (HKUST). I am now working towards the M.Phil. degree at the same department, under the supervision of Prof. Oscar C. Au.

Currently, I am also a research associate under Prof. Fernando De la Torre at the Human Sensing Lab, Robotics Institute, Carnegie Mellon University. Between Nov. 2010 and Aug. 2011, I was an intern student at the Shenzhen Key Laboratory for Computer Vision and Pattern Recognition directed by Prof. Xiaoou Tang in Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, where I was co-supervised by Prof. Jianzhuang Liu and Dr. Chunjing Xu.

My research interests mainly focus on Clustering Structure Analysis, Image Segmentation, Mode Seeking, Metric & Manifold Learning and Object Detection.

I am the awardee of  the 2009-2010 HKTIIT Post-Graduate Excellence Scholarships.

Educational Background

Selected Publications

Journal Papers

Conference Papers

Selected Projects

Tree Embedded Mode Seeking for Manifold Structured Data Clustering and Image Segmentation

We present a novel framework for tree-structure embedded density estimation and its fast approximation for mode seeking. Given any undirected, connected and weighted graph, the density function is defined as a joint representation of the feature space and the distance domain on the graph’s spanning tree. Since the distance domain of a tree is a constrained one, mode seeking can not be directly achieved by traditional mean shift in both domain. we address this problem by introducing node shifting with force competition and its fast approximation. The new formulation of this problem can lead to many advantages and new characteristics in its application.

This work appears in CVPR 2011. Project page under construction...

Automatic Object Segmentation from Large Scale 3D Urban Point Clouds


 

We present a system that can automatically segment objects in large scale 3D point clouds obtained from urban ranging images. The system consists of three steps: The first one involves a ground detection process that can detect relatively complex terrain and separate it from other objects. The second step superpixelizes the remaining objects to speed up the segmentation process. In the final step, a manifold embedded mode seeking method is adopted to segment the point clouds. Even though the segmentation of urban objects is a challenging problem in terms of accuracy and problem scale, our system can efficiently generate very good segmentation results. The proposed manifold learning effectively improves the segmentation performance due to the fact that continuous artificial objects often have manifold-like structures.

This work appears in ACM-MM 2011. Project page under construction...

Automatic Real-time Car Detection with Boosted cascade of Harr-like features and HoG

Due to the limited description ability of Haar features, the detection accuracy decreases drastically in some cases where the objects are not that regular like human face in appearance. We aim at finding a balance between the detection accuracy and speed in establishing an automatic vehicle detection system for crossroad surveillance cameras. The proposed approach combines SVM HoG detection and boosted cascades. The fast rejection of the most candidates in the first few stages of cascades using Haar features provides an opportunity to greatly reduce the work load of subsequent classifications which are more precise but requires much larger computation power.

Result video of the system is available here.

ELEC 547 Convex Optimization Course Project: Mode Seeking with Convex Shift and Interactive Learning

  
 

  

We give a brief introduction of nonparametric density estimation and the family of mode seeking methods under this framework. Regarding the limit of current methods, we propose several improvements. The major contribution is basically two fold: 1. We show that for mean shift with an Epanechnikov kernel, each kernel shift can actually be formulated in a convex form. By doing this we further generalize it for several convex metrics - e.g., KL divergence and Jeffrey divergence - where it is difficult or impossible to clearly define a ”mean” for mode seeking. 2. We also introduce interactive learning to image segmentation which is a specific application of mode seeking algorithms. By adopting interactive segmentation through convex learning, better segmentation results can be achieved over segmentation with traditional mode seeking algorithms.

The project report is available here.

Honors and Awards

Courses Learned (Click to see the list of courses)

More Information (Click to get more info.)

Friends & Collaborators

 

This page has been visited

 电影网站流量统计

times

 
Locations of visitors to this page