Temporal segmentation and assignment of successive actions in a long-term video
Temporal segmentation and assignment of successive actions in a long-term video
复制标题
DOI：
10.1016/j.patrec.2012.10.023
复制
发表时间：
2013-11
期刊：
影响因子：
--
通讯作者：
Guoliang Lu;Mineichi Kudo;J. Toyama
中科院分区：
文献类型：
--
作者：
Guoliang Lu;Mineichi Kudo;J. Toyama
长期视频序列中连续动作的时间分割一直是计算机视觉中长期存在的问题。在本文中，我们开发了一种新颖的基于学习的框架。给定一个视频序列，所提出的选择算法仅选择少数特征帧，然后以成对的方式计算训练模型的似然，最后获得分割作为最优模型序列以实现最大似然。 IXMAS 数据集在帧级别的平均准确率达到 80.5%，仅使用所有帧的 16.5%，每个视频的计算时间为 1.57 秒，平均有 1160 帧。
Temporal segmentation of successive actions in a long-term video sequence has been a long-standing problem in computer vision. In this paper, we exploit a novel learning-based framework. Given a video sequence, only a few characteristic frames are selected by the proposed selection algorithm, and then the likelihood to trained models is calculated in a pair-wise way, and finally segmentation is obtained as the optimal model sequence to realize the maximum likelihood. The average accuracy on IXMAS dataset reached to 80.5% at frame level, using only 16.5% of all frames in computation time of 1.57 s per video which has 1160 frames on the average.