SteROI-D: System Design and Mapping for Stereo Depth Inference on Regions of Interest

TL;DR

SteROI-D is an energy-efficient stereo depth inference system for AR/VR devices that leverages region-of-interest and temporal sparsity, achieving significant energy savings through a novel mapping methodology.

Contribution

The paper presents SteROI-D, a system that exploits ROI and temporal sparsity with a systematic mapping approach to reduce energy consumption in stereo depth estimation.

Findings

Achieves up to 4.35x energy reduction on a 28nm prototype.

Supports diverse ROIs with a flexible compute fabric.

Effectively handles dynamic ROIs for energy efficiency.

Abstract

Machine learning algorithms have enabled high quality stereo depth estimation to run on Augmented and Virtual Reality (AR/VR) devices. However, high energy consumption across the full image processing stack prevents stereo depth algorithms from running effectively on battery-limited devices. This paper introduces SteROI-D, a full stereo depth system paired with a mapping methodology. SteROI-D exploits Region-of-Interest (ROI) and temporal sparsity at the system level to save energy. SteROI-D's flexible and heterogeneous compute fabric supports diverse ROIs. Importantly, we introduce a systematic mapping methodology to effectively handle dynamic ROIs, thereby maximizing energy savings. Using these techniques, our 28nm prototype SteROI-D design achieves up to 4.35x reduction in total system energy compared to a baseline ASIC.