Jun Zhu

Jun Zhu portrait 

Jun Zhu, Ph.D.
Postdoctoral Research Fellow
Department of Statistics, University of California, Los Angeles

Email: jzh at ucla dot edu
Address: Room 2851, Geology Building, UCLA, Los Angeles, CA 90095
Tel: +1-310-866-7087

Curriculum Vitae

About me

I am a postdoctoral research fellow working with Prof. Alan Yuille at UCLA. I received the Ph.D. degree at Department of Electronic Engineering, Shanghai Jiao Tong University in 2013.

My research interests include computer vision, machine learning, image categorization, scene recognition, object detection, and human action recognition. Particularly, I am devoted to building hierarchical compositional models (e.g., the "tangram model") for scene represeantaion and recognition, and deep part-based models for object and semantic part detection. Besides, I devolopped computational models and algorithms for weakly-supervised semantic segmentation and human action recognition.

During my Ph.D., I was a research intern at Microsoft Research Asia, working with Prof. Zhuowen Tu and Dr. Baoyuan Wang. I have also been a visiting student for one year in Lotus Hill Institue of Computer Vision and Information Science, under supervision of Prof. Song-Chun Zhu.


Research Highlight

Deep Part-Based Models 

DeePM: A Deep Part-Based Model for Object Detection and Semantic Part Localization
Jun Zhu, Xianjie Chen and Alan Yuille
[Arxiv] [Bibtex]

Pose Guided Human Parsing 

Pose-Guided Human Parsing by an AND/OR Graph Using Pose-Context Features
Fangting Xia, Jun Zhu, Peng Wang* and Alan L. Yuille (*: equal contribution with the second author)
to appear in The Thirtieth AAAI Conference on Articial Intelligence (AAAI-16), 2016 (Oral).
[Arxiv] [PDF] [Bibtex]

Expectation Loss SVM (e-SVM) 

Learning From Weakly Supervised Data by The Expectation Loss SVM (e-SVM) Algorithm
Jun Zhu, Junhua Mao and Alan L. Yuille
in Conference on Advances in Neural Information Processing Systems (NIPS), 2014.
[PDF] [Bibtex]

Tangram Model 

A Reconfigurable Tangram Model for Scene Representation and Categorization
Jun Zhu, Tianfu Wu, Song-Chun Zhu, Xiaokang Yang and Wenjun Zhang
IEEE Transaction on Image Processing (TIP).
[PDF] [Dataset] [Bibtex]

Learning Reconfigurable Scene Representation by Tangram Model
Jun Zhu, Tianfu Wu, Song-Chun Zhu, Xiaokang Yang and Wenjun Zhang
in IEEE Workshop on the Applications of Computer Vision (WACV), 2012.
[PDF] [Poster] [Code] [Bibtex]

[Project Page]

Action Recognition with Actons 

Action Recognition with Actons
Jun Zhu, Baoyuan Wang, Xiaokang Yang, Wenjun Zhang and Zhuowen Tu
in IEEE International Conference on Computer Vision (ICCV), 2013.
[PDF] [Poster] [Supplementary Material] [Bibtex]

A Two-Layer Representation for Large-Scale Action Recognition
Jun Zhu, Baoyuan Wang, Xiaokang Yang, Wenjun Zhang and Zhuowen Tu
in ICCV Workshop on Action Recognition with a Large Number of Classe (THUMOS), 2013.

Hierarchical Spatial Pooling 

Image Classification by Hierarchical Spatial Pooling with Partial Least Squares Analysis
Jun Zhu, Weijia Zou, Xiaokang Yang, Rui Zhang, Quan Zhou and Wenjun Zhang
in British Machine Vision Conference (BMVC), 2012.
[PDF] [Abstract] [Code] [Bibtex]

[Project Page]

Selected Conference Papers

Journal Papers

Note: "All materials are presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors and other holders (publishers)".

Academic Services

Journal Reviewers Conference Reviewers