Ajou University repository

Improving the Efficiency of CMOS Image Sensors through In-Sensor Selective Attention
  • Zhang, Tianyi ;
  • Kasichainula, Kishore ;
  • Jee, Dong Woo ;
  • Yeo, Injune ;
  • Zhuo, Yaoxin ;
  • Li, Baoxin ;
  • Seo, Jae Sun ;
  • Cao, Yu
Citations

SCOPUS

0

Citation Export

Publication Year
2023-01-01
Journal
Proceedings - IEEE International Symposium on Circuits and Systems
Publisher
Institute of Electrical and Electronics Engineers Inc.
Citation
Proceedings - IEEE International Symposium on Circuits and Systems, Vol.2023-May
Keyword
image sensorlatencyobject detectionpower consumptionsaliency detectionselective attention
Mesh Keyword
% reductionsCMOS image sensorHuman visionLatencyObjects detectionProcessing stepsSaliency detectionSalient objectsSelective attentionSelective attention mechanism
All Science Classification Codes (ASJC)
Electrical and Electronic Engineering
Abstract
Inspired by the selective attention mechanism in human vision, we propose to introduce a saliency-based processing step in the CMOS image sensor, to continuously select pixels corresponding to salient objects and feedback such information to the sensor, instead of blindly passing all pixels to the sensor output. To minimize the overhead of saliency detection in this feedback loop, we propose two techniques: (1) saliency detection with low-precision, down-sampled grayscale images, and (2) Optimization of the loss function and model structure. Finally, we pad the minimum number of pixels around the selected pixels to maintain the accuracy of object detection (OD). Our method is experimented with two types of OD algorithms on three representative datasets. At the similar OD accuracy with the full image, our proposed selective feedback method successfully achieves 70.5% reduction in the volume of output pixels for BDD100K, which translates to 4.3× and 3.4× reduction in power consumption and latency, respectively.
ISSN
0271-4310
Language
eng
URI
https://aurora.ajou.ac.kr/handle/2018.oak/36972
https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85167658524&origin=inward
DOI
https://doi.org/10.1109/iscas46773.2023.10181835
Type
Conference
Show full item record

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Related Researcher

 Jee, Dong Woo Image
Jee, Dong Woo지동우
Department of Electrical and Computer Engineering
Read More

Total Views & Downloads

File Download

  • There are no files associated with this item.