| Paper: | WA-P4.2 |
| Session: | Object Recognition I / Interpolation and Superresolution |
| Time: | Wednesday, September 19, 09:50 - 12:30 |
| Presentation: |
Poster
|
| Title: |
OBJECT RECOGNITION BASED ON DEPENDENT PACHINKO ALLOCATION MODEL |
| Authors: |
Yuanning Li; Institute of Computing Technology, Chinese Academy of Sciences | | |
| | Weiqiang Wang; Chinese Academy of Sciences | | |
| | Wen Gao; Institute of Computing Technology, Chinese Academy of Sciences | | |
| Abstract: |
Recently“bag of words” model becomes popular in the approaches to object recognition. These approaches model an image as a collection of local patches called “visual words”, and recognize objects through topics inferring. In this paper, we apply an extension of Pachinko Allocation Model(PAM) to object recognition. Our approach models the topics correlation in a hierarchical structure. To relax the independent assumption for visual words and refine the topic inferring, we incorporate the prior knowledge of co-occurrence dependence among visual words into PAM. Highly competitive recognition results on Caltech4 and Caltech101 datasets show our approach is more expressive and discriminative than most existing methods of object recognition. |
