Reconfiguring the Imaging Pipeline for Computer Vision

TL;DR

This paper proposes a reconfigurable imaging pipeline that switches between high-quality photography and low-power vision modes, significantly reducing energy consumption with minimal impact on vision accuracy.

Contribution

It introduces a configurable imaging pipeline and a sensor design that enables energy-efficient vision modes by skipping certain ISP stages.

Findings

Disabling ISP stages like demosaicing and gamma compression has minimal impact on vision accuracy.

The proposed sensor design with adjustable resolution and tunable ADCs enables 75% energy savings.

Vision mode reduces energy consumption while maintaining acceptable task performance.

Abstract

Advancements in deep learning have ignited an explosion of research on efficient hardware for embedded computer vision. Hardware vision acceleration, however, does not address the cost of capturing and processing the image data that feeds these algorithms. We examine the role of the image signal processing (ISP) pipeline in computer vision to identify opportunities to reduce computation and save energy. The key insight is that imaging pipelines should be designed to be configurable: to switch between a traditional photography mode and a low-power vision mode that produces lower-quality image data suitable only for computer vision. We use eight computer vision algorithms and a reversible pipeline simulation tool to study the imaging system's impact on vision performance. For both CNN-based and classical vision algorithms, we observe that only two ISP stages, demosaicing and gamma compression, are critical for task performance. We propose a new image sensor design that can compensate for skipping these stages. The sensor design features an adjustable resolution and tunable analog-to-digital converters (ADCs). Our proposed imaging system's vision mode disables the ISP entirely and configures the sensor to produce subsampled, lower-precision image data. This vision mode can save ~75% of the average energy of a baseline photography mode while having only a small impact on vision task accuracy.

Figures

Full text

See pages 1-last of archive.pdf