Abstract | ||
---|---|---|
Face detection and alignment are highly-correlated, computation-intensive tasks, without being flexibly supported by any facial-oriented accelerator yet. This work proposes the first unified accelerator for multi-face detection and alignment, along with the optimizations on multi-task cascaded convolutional networks algorithm, to implement both multi-face detection and alignment. First, the clustering non-maximum suppression is proposed to significantly reduce intersection over union computation and eliminate the hardware-interfer-ence sorting process, bringing 16.0% speed-up without any loss. Second, a new pipeline architecture is presented to implement the proposal network in more computation-efficient manner, with 41.7% less multiplier usage and 38.3% decrease in memory capacity compared with the similar method. Third, a batch schedule mechanism is proposed to improve hardware utilization of fully-connected layer by 16.7% on average with variable input number in batch process. Based on the TSMC 28 nm CMOS process, this accelerator only consumes 6.7ms at 400 MHz to simultaneously process 5 faces for each image and achieves 1.17 TOPS/W power efficiency, which is 54.8× higher than the state-of-the-art solution.
|
Year | DOI | Venue |
---|---|---|
2019 | 10.1145/3316781.3317736 | Proceedings of the 56th Annual Design Automation Conference 2019 |
Keywords | Field | DocType |
multiface detection,multiface dalignment,facial-oriented accelerator,unified accelerator,clustering nonmaximum suppression,hardware-interference sorting process,batch schedule mechanism,TSMC CMOS process,multitask cascaded convolutional network algorithm,union computation,pipeline architecture,memory capacity,computation-intensive tasks,size 28.0 nm,time 6.7 ms,frequency 400.0 MHz | Electrical efficiency,Computer science,Real-time computing,Multiplier (economics),Sorting,Batch processing,Face detection,Computer hardware,Cluster analysis,TOPS,Computation | Conference |
ISSN | ISBN | Citations |
0738-100X | 978-1-4503-6725-7 | 1 |
PageRank | References | Authors |
0.36 | 8 | 8 |
Name | Order | Citations | PageRank |
---|---|---|---|
Huiyu Mo | 1 | 8 | 3.59 |
leibo liu | 2 | 816 | 116.95 |
Wenping Zhu | 3 | 22 | 6.59 |
Qiang Li | 4 | 599 | 54.40 |
Hong Liu | 5 | 301 | 26.44 |
Wenjing Hu | 6 | 11 | 6.39 |
Yao Wang | 7 | 1 | 0.36 |
Shaojun Wei | 8 | 555 | 102.32 |