.png){kind=icon}
.png){kind=icon}
Return
AnchorCapsule: A Datastream-Serving Post-Processor for Object Detection in Embedded Vision SoC
.png){kind=icon}
PRE
AI
DOI：10.1109/TCSII.2022.3200038.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
We present AnchorCapsule, a post -processor for object detection to relieve the workload of CPUs with minimal cost in latency. The capsule takes advantage of a 'datastreamserving' architecture paradigm, where computing logic is built to cater for the throughput of DRAM datastreams and simultaneously minimize the usage of on -chip buffers. The implementation results show that AnchorCapsule has a tiny area of 0.038mm(2) but achieves 1 order of magnitude faster than Intel Xeon and 2 orders of magnitude faster than ARM A53, resulting in an endto -end system -level latency reduction of 46.7% and 16.7% for Yolo-v3-tiny and Yolo-v3 networks, respectively. The precision of AnchorCapsule has been validated on 1,500 samples from three prestige datasets, giving a promising result of 98%+ accuracy in the bounding box (BBox) coordinates, 99%+ in BBox sizes and 100% in object types. Compared with state-of-the-art, AnchorCapsule can filter 9,408 candidate BBoxes in a single run, which is 3.5x in the processing capacity of best-known published work.
Keywords:
Object detection
post-processing
datastream-serving architecture
Journal
I
IF:
4.9
Papers:
8.8K
Citations:
2.5W