Automatic parsing of ball-sports video sequences and its application to soccer
This thesis describes the study, proposal and design of a video model using a priori knowledge to parse generic ball-sports video sequences and automatically classify them into proper play categories based on their contextual information. An automatic video parser has been implemented using the prop...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Theses and Dissertations |
| Published: |
2010
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/39002 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Summary: | This thesis describes the study, proposal and design of a video model using a priori knowledge to parse generic ball-sports video sequences and automatically classify them into proper play categories based on their contextual information. An automatic video parser has been implemented using the proposed video model on a specific ball-sports game - soccer. Extensive experimental results showed that our proposed video model is effective. Tests on 720 soccer video clips of 18 play categories have a recognition rate of 87.7%. |
|---|