| Objective Performance Evaluation of Video Segmentation Algorithms with Ground-Truth |
| |
| 作者姓名： | 杨高波 张兆扬 |
| |
| 作者单位： | ScholofCommunicationandInformationEngineering,ShanghaiUniversity,Shanghai200072,P.R.China |
| |
| 基金项目： | theNationalNatureScienceFoundationofChina (GrantNo .60 172 0 2 0 ) |
| |
| 摘 要： | While the development of particular video segmentation algorithms has attracted considerable research interest, relatively little effort has been devoted to provide a methodology for evaluating their performance. In this paper, we propose a methodology to objectively evaluate video segmentation algorithm with ground-truth, which is based on computing the deviation of segmentation results from the reference segmentation. Four different metrics based on classification pixels, edges, relative foreground area and relative position respectively are combined to address the spatial accuracy. Temporal coherency is evaluated by utilizing the difference of spatial accuracy between successive frames. The experimental results show the feasibility of our approach. Moreover, it is computationally more efficient than previous methods. It can be applied to provide an offline ranking among different segmentation algorithms and to optimally set the parameters for a given algorithm.
|
| 关 键 词： | 视频分割算法 目标性能评估 图像处理 图像编码 |
| 收稿时间： | 14 April 2003 |
Objective performance evaluation of video segmentation algorithms with ground-truth |
| |
| Yang?Gao-bo,Zhang?Zhao-yang.Objective Performance Evaluation of Video Segmentation Algorithms with Ground-Truth[J].Journal of Shanghai University(English Edition),2004,8(1):70-74. |
| |
| Authors: | Yang Gao-bo Zhang Zhao-yang |
| |
| Institution: | (1) Schol of Communication and Information Engineering, Shanghai University, 200072 Shanghai, P.R. China |
| |
| Abstract: | While the development of particular video segmentation algorithms has attracted considerable research interest, relatively little effort has been devoted to provide a methodology for evaluating their performance. In this paper, we propose a methodology to objectively evaluate video segmentation algorithm with ground-truth, which is based on computing the deviation of segmentation results from the reference segmentation. Four different metrics based on classification pixels, edges, relative foreground area and relative position respectively are combined to address the spatial accuracy. Temporal coherency is evaluated by utilizing the difference of spatial accuracy between successive frames. The experimental results show the feasibility of our approach. Moreover, it is computationally more efficient than previous methods. It can be applied to provide an offline ranking among different segmentation algorithms and to optimally set the parameters for a given algorithm. |
| |
| Keywords: | video object segmentation performance evaluation MPEG-4 |
| 本文献已被 CNKI 维普 SpringerLink 等数据库收录！ |
