Efficient Hand Articulations Tracking using Adaptive Hand Model and Depth map

摘要

Real-time hand articulations tracking is important for many applications suchas interacting with virtual / augmented reality devices or tablets. However,most of existing algorithms highly rely on expensive and high power-consumingGPUs to achieve real-time processing. Consequently, these systems areinappropriate for mobile and wearable devices. In this paper, we propose anefficient hand tracking system which does not require high performance GPUs. Inour system, we track hand articulations by minimizing discrepancy between depthmap from sensor and computer-generated hand model. We also initialize hand poseat each frame using finger detection and classification. Our contributions are:(a) propose adaptive hand model to consider different hand shapes of userswithout generating personalized hand model; (b) improve the highly efficientframe initialization for robust tracking and automatic initialization; (c)propose hierarchical random sampling of pixels from each depth map to improvetracking accuracy while limiting required computations. To the best of ourknowledge, it is the first system that achieves both automatic hand modeladjustment and real-time tracking without using GPUs.