1 Hit in 2.4 sec

CoDeNet: Efficient Deployment of Input-Adaptive Object Detection on Embedded FPGAs [article]

Zhen Dong, Dequan Wang, Qijing Huang, Yizhao Gao, Yaohui Cai, Tian Li, Bichen Wu, Kurt Keutzer, John Wawrzynek
2021 arXiv   pre-print
In this work, we harness the flexibility of FPGAs to develop a novel object detection pipeline with deformable convolutions.  ...  With our high-efficiency implementation, our solution reaches 26.9 frames per second with a tiny model size of 0.76 MB while achieving 61.7 AP50 on the standard object detection dataset, Pascal VOC.  ...  We devise several configurations of CoDeNet to facilitate the latency-accuracy tradeoffs for our final object detection solution on the embedded FPGAs.  ... 
arXiv:2006.08357v2 fatcat:lhcthpyf5jb2jihvzvjmpgyvbe