A real-time 17-scale object detection accelerator with adaptive 2000-stage classification in 65nm CMOS

摘要

This paper presents an object detection accelerator that features many-scale (17), many-object (up to 50), multi-class (e.g., face, traffic sign), and high accuracy (average precision (AP) of 0.81/0.72 for AFW/BTSD datasets) detection. Employing 10 gradient/color channels, integral features are extracted and 2,000 simple classifiers for rigid boosted templates are adaptively combined to make a strong classification. The prototype chip implemented in 65nm CMOS demonstrates 16-40 frames per second and 22-160 mW power at 0.6-1.0V supply.