Abstract: Real-time detection and tracking of gesture based on computer vision is key technology in the human-computer interaction, many algorithms for detection and tracking of gesture have been investigated in recent decades, which consisted of detection and tracking algorithm as two independent modules. These algorithms are not applicable because they are easily impacted by the gesture posture, object occlusion, motion blur and external environment etc. In this paper, a novel algorithm for detection and tracking of gesture is proposed based on real time gesture detection and tracking, the main advantage compared with the traditional algorithm is that this proposed algorithm effectively fuses the results from detection and tracking modules. The proposed algorithm can initialize the gesture automatically and can recover by itself when the gesture is dropped. Experimental results show that the proposed algorithm can achieve rapid, accurate and continuous recognition of the gesture motion and meet the requirements of the human-computer interaction.

Key words: compressive sensing, gesture detection, gesture tracking

摘要： 基于计算机视觉的实时手势检测与跟踪算法是人机交互领域的一项关键技术，传统的手势检测与跟踪算法将检测和跟踪分成两个独立的模块进行，检测与跟踪结果受手势姿态变化、目标遮挡、运动模糊以及外界环境干扰等因素的影响。提出了一种基于压缩感知的实时手势检测和跟踪算法，将基于检测得到的手势信息与基于压缩感知跟踪算法得到的目标信息进行有效融合，从而实现有效的手势检测与跟踪，与传统算法相比，该算法能实现手势跟踪自动初始化和跟踪错误后自我恢复功能。实验结果表明，提出的算法能对手势运动进行快速、连续、准确的识别，满足人机交互的要求。

关键词: 压缩感知, 手势识别, 手势跟踪