出版社:Academy & Industry Research Collaboration Center (AIRCC)
摘要:The paper introduces a case study of design and implementation of Intel-sponsored real-timeface detection system conducted in University of Michigan—Shanghai Jiao Tong UniversityJoint Institute (JI). This work is teamed up totally by 15 JI students and developed in threephases during 2013 and 2014. The system design of face detection is based on Intel HighDefinition (HD) 4000 graphics and OpenCL. With numerous techniques including theaccelerated pipeline over CPU and GPU, image decomposition, two-dimensional (2D) taskallocation, and the combination of Viola-Jones algorithm and continuously adaptive mean-shift(Camshift) algorithm, the speed reaches 32 fps for real-time multi-view face detection. Plus, thefrontal view detection accuracy obtains 81% in Phase I and reaches 95% for multi-viewdetection, in Phase III. Furthermore, an innovative application called face-detection gamecontroller (FDGC) is developed. At the time of this writing, the technology has beenimplemented in wearable devices and mobile with Intel cores.
关键词:Multi-view face detection; real-time; graphic processing unit (GPU); OpenCL; face-detection;game controller (FDGC).