Ocellus: Highly Parallel Convolution-in-Pixel Scheme Realizing Power-Delay-Efficient Edge Intelligence
Document Type
Conference Proceeding
Publication Date
1-1-2023
Abstract
With the advent of Edge Intelligence (EI) devices, always-on intelligent and self-powered visual perception systems are receiving considerable attention. These emerging systems require continuous sensing and instant processing; however, the high energy data conversion/transmission of raw data and the limited available energy and computation resources make designing energy-efficient and low bandwidth CMOS vision sensors vital but challenging. This paper proposes a low-power integrated sensing and computing engine, namely Ocellus, which considerably decreases power costs of data movement/conversion and enables data/compute -intensive neural network tasks. Ocellus offers several unique features, including a highly parallel analog convolution-in-pixel scheme and reconfigurable filtering modes with filter pruning capability. These features realize low-precision ternary weight neural networks to mitigate the overhead of analog-to-digital converters and analog buffers. Moreover, the proposed structure supports a zero-skipping scheme to further reduce power consumption. Our circuit-to-application cosimulation results demonstrate comparable, even better, accuracy to the full-precision baseline on object classification tasks, while it achieves a frame rate of 1000 and efficiency of 1.45 TOp/s/W.
Identifier
85173078497 (Scopus)
ISBN
[9798350311754]
Publication Title
Proceedings of the International Symposium on Low Power Electronics and Design
External Full Text Location
https://doi.org/10.1109/ISLPED58423.2023.10244476
ISSN
15334678
Volume
2023-August
Grant
2216772
Fund Ref
National Science Foundation
Recommended Citation
Tabrizchi, Sepehr; Angizi, Shaahin; and Roohi, Arman, "Ocellus: Highly Parallel Convolution-in-Pixel Scheme Realizing Power-Delay-Efficient Edge Intelligence" (2023). Faculty Publications. 2248.
https://digitalcommons.njit.edu/fac_pubs/2248