融合协作表达和时空金字塔的人体行为识别

摘要

为了能够充分正确地利用视频数据从而保证直方图向量对行为的高描述性,提出了一种新颖、有效的人体行为识别算法.所提算法融合了协作表达方法和时空金字塔表示,并且采用引入局部限制的方式保持了局部特征与其相邻字典向量之间的相似性.协作表达的引入使得算法可以得到一个闭式,避免了在稀疏表示分类算法中因迭代造成的高时间复杂度问题,同时为了增强直方图向量的行为鉴别能力,算法巧妙地借助时空金字塔描述视频中特征的结构化分布.在KT H和Hollyw ood2人体动作数据集上的识别率分别可达96%和59.1%,明显高于其他对比算法.%Human behavior description is the key problem in behavior recognition .In order to make full use of video data and ensure the highly descriptive histogram vector of behavior ,a novel and effective algorithm for human behavior recognition is proposed . The proposed algorithm combining collaborative representation with space-time pyramid ,which preserves the similarity between the test sample and its neighboring training samples by employing the locality . The introduction of cooperative expression allow s the algorithm to obtain a closed-form solution . Therefore ,the algorithm not only avoids the dilemma of expensive computations in the sparse representation , but also improves the discriminative pow er of interest points by encoding the structured distribution of features in the video . The recognition rates on the KTH and Hollywood2 human action datasets reach 96% and 59 .1% ,which is better than the state-of-the-art methods .