A Compression-Compilation Co-Design Framework Towards Real-Time Object Detection on Mobile Devices

The rapid development and wide utilization of object detection techniques have aroused requirements for both accuracy and speed of object detectors. In this work, we propose a compression-compilation co-design framework to achieve real-time YOLOv4 inference on mobile devices. We propose a novel fine-grained structured pruning, which maintain high accuracy while achieving high hardware parallelism. Our pruned YOLOv4 achieves 48. 9 mAP and 17 FPS inference speed on an off-the-shelf Samsung Galaxy S20 smartphone, which is 5. 5× faster than the original state-of-the-art detector YOLOv4.

Authors

• Yuxuan Cai
• Geng Yuan University of Georgia
• Hongjia Li School of Medical Technology, Beijing Institute of Technology
• Wei Niu University of Georgia
• Yanyu Li Northeastern University
• Xulong Tang
• Bin Ren College of William & Mary
• Yanzhi Wang Northeastern University

Keywords

No keywords are indexed for this paper.