Appearance-based Person Model for Multiple People Tracking
碩士 === 國立東華大學 === 資訊工程學系 === 94 === Getting depth information from stereo images is a conventional problem in computer vision. Recently, people detection and tracking by stereo cameras has become popular due the advance in computation power and the efficiency of depth-by-stereo algorithms. Typically...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | zh-TW |
| Published: |
2006
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/55524878108587856578 |
| Summary: | 碩士 === 國立東華大學 === 資訊工程學系 === 94 === Getting depth information from stereo images is a conventional problem in computer vision. Recently, people detection and tracking by stereo cameras has become popular due the advance in computation power and the efficiency of depth-by-stereo algorithms.
Typically, two kinds of information can be used for person detection and tracking. The geometry information is ideal for tracking and the appearance information is critical for recognition. In this paper, we propose a system that not only detects people based on the color and depth information, but also tracks/recognizes people by integrating the geometry and appearance information.
|
|---|