Paper Info
Title
VOTA: A 2.45TFLOPS/W Heterogeneous Multi-Core Visual Object Tracking Accelerator Based on Correlation Filters
Abstract
VOTA is a domain-specific accelerator for correlation filter (CF)-based visual object tracking (VOT). It encompasses a Winograd convolution core, a FFT core and a vector core in a high-bandwidth starring topology. VOTA’s frame-based instructions and execution enable a 537GFLOPS performance and reduce the code size. An instruction-chaining mechanism permits inter-core pipelining to improve the utilization to 84.2%. A 10.2mm <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> 28nm FP16 VOTA prototype incorporating a RISC-V host CPU is measured to achieve 2.45TFLOPS/W at 0.72V. Running OPCF, a CF-based VOT enhanced by adaptive boosting and particle filters, the chip achieves 1157FPS on 640×480 input frames at 0.9V and 175MHz, consuming 296mW.
Year
DOI
Venue
2021
10.23919/VLSICircuits52068.2021.9492379
2021 Symposium on VLSI Circuits
Keywords
DocType
ISSN
domain-specific accelerator,correlation filter-based visual object tracking,Winograd convolution core,FFT core,vector core,instruction-chaining mechanism,inter-core pipelining,CF-based VOT,adaptive boosting,particle filters,eterogeneous multicore visual object tracking accelerator,high-bandwidth star-ring topology,VOTA frame-based instructions,RISC-V host CPU,size 28.0 nm,voltage 0.72 V,voltage 0.9 V,power 296.0 mW,frequency 175.0 MHz,computer speed 537.0 GFLOPS
Conference
2158-5601
ISBN
Citations 
PageRank 
978-1-6654-4766-9
1
0.35
References 
Authors
0
6
Name
Order
Citations
PageRank
Junkang Zhu110.35
Wei Tang210.69
Ching-En Lee3215.86
Haolei Ye410.69
Eric McCreath510.69
Zhengya Zhang650248.41