-
摘要: 视频目标分割是视频监视与视频目标跟踪、视频目标识别以及视频编辑的基础. 本文提出了一种基于隐条件随机场 (Hidden conditional random fields, HCRF) 的自适应视频分割算法, 利用 HCRF 模型对视频序列中的时空邻域关系建模. 使用在线学习的方式对相应的参数进行调整, 实现对时空邻域约束关系的权重调整, 提高视频目标分割细节上的效果. 大量的数据测试表明, 与高斯混合模型 (Gaussian mixture model, GMM) 和联合时空的马尔可夫随机场 (Markov random fields, MRF) 等算法相比, 该算法的分割错误率分别降低了23\%和19\%.Abstract: Video object segmentation is important for video surveillance and video object tracking, video object recognition and video editing. An adaptive video segmentation algorithm based on hidden conditional random fields (HCRFs) is proposed, which models spatio-temporal constraints of video sequence. In order to improve the segmentation quality, the weights of spatio-temporal constraints are adaptively updated by on-line learning of HCRFs. The experimental results have demonstrated that the error ratio of our algorithm is reduced by 23\% and 19\%, respectively, compared with Gaussian mixture model (GMM) and spatio-temporal Markov random fields (MRF).
-
Key words:
- Video segmentation /
- hidden conditional random fields /
- on-line learning
-
计量
- 文章访问数: 2559
- HTML全文浏览量: 106
- PDF下载量: 1387
- 被引次数: 0
下载:
