Far East Journal of Experimental and Theoretical Artificial Intelligence
Volume 4, Issue 2, Pages 73 - 87
(November 2009)
|
|
A FIGURE EXTRACTION AND SYNTHESIS SYSTEM USING LEARNING VECTOR QUANTIZATION NEURAL NETWORKS
Chuan-Yu Chang (Taiwan) and Zong-Yu Tsai (Taiwan)
|
Abstract:
Extracting complete figures from videos in complicated environments is difficult. A novel figure extraction and synthesis system with capability of extracting figures from consecutive frames in a messy environment is proposed in this paper. A figure template is constructed based on the face detection results and some image processing techniques. Features of figure and background are extracted from the figure images. By means of these features, a learning vector quantization neural network (LVQNN) is applied to classify the uncertain regions into figural and non-figural objects. The extracted figure can be further synthesized into an optional cinestrip. Experimental results show the proposed method successfully extracts the figure object from a complex background environment. |
| Keywords and phrases: moving object extraction, human shape extraction, learning vector quantization neural networks. |
| Communicated by Shun-Feng Su |
| Number of Downloads: 150 | Number of Views: 502 |
|
