Unsupervised video segmentation and its application to activity recognition

Cheng, Hsien Ting

Permalink

https://hdl.handle.net/2142/72891 Copy

Description

Title

Unsupervised video segmentation and its application to activity recognition

Author(s)

Cheng, Hsien Ting

Issue Date

2015-01-21

Director of Research (if dissertation) or Advisor (if thesis)

Ahuja, Narendra

Doctoral Committee Chair(s)

Ahuja, Narendra

Committee Member(s)

• Forsyth, David A.
• Hasegawa-Johnson, Mark A.
• Huang, Thomas

Department of Study

Electrical & Computer Eng

Discipline

Electrical & Computer Engr

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• segmentation
• Video segmentation
• Unsupervised clustering
• Activity recognition
• Multiple instance learning

Abstract

We addressed the fundamental problem of computer vision: segmentation and recognition, in the space-time domain. With the knowledge that generic image segmentation introduces unstable regions due to illumination, com- pression, etc., we utilized temporal information to achieve consistent 3D video segmentation. By exploiting non-local structure in both spatial and temporal space, the instabilities of the segmented regions were alleviated. A segmentation tree was built within every frame, and the label consistency was enforced within each subtree (i.e. spatial clique). By roughly tracking 2D regions across each frame, temporal clique was built in which label consis- tency was enforced as well. The high-order (more than binary) Conditional Random Field (CRF) is designed and solved efficiently. Experimental results demonstrate high-quality segmentation quantitatively and qualitatively. Taking segmented 3D regions, called tubes, as input, we developed an activity recognition framework not only to determine which activity existed in a video but also to locate where it happens. A robust tube feature was extracted with photometric and shape dynamics information. Activity was described as a Parts Activity Model (PAM) with a root template and four- part template under the root. Given the nature of the activity recognition problem that only some parts on the video were used to determine the activity label, we used Multiple Instance Learning (MIL) to formulate the problem. Latent variables included a tube index and the parts location under the root template. Experiments were conducted on three well-known datasets and a state-of-the-art result was achieved.

Graduation Semester

2014-12

Permalink

http://hdl.handle.net/2142/72891

Copyright and License Information

Copyright 2014 Hsien Ting Cheng

Owning Collections