Analysis & Computational Complexity Reduction of Monocular and Stereo Depth Estimation Techniques

TL;DR

This paper investigates reducing computational complexity in monocular and stereo depth estimation methods, demonstrating significant model size reductions with minimal accuracy loss for real-time autonomous system applications.

Contribution

The study presents methods to significantly reduce model sizes of monocular and stereo depth estimation techniques while maintaining high accuracy, enabling more efficient real-time deployment.

Findings

Monocular depth model size reduced by ~75% with less than 2% accuracy loss.

Stereo depth model size reduced by ~20% with less than 3% accuracy loss.

Smaller models can perform competitively in depth estimation tasks.

Abstract

Accurate depth estimation with lowest compute and energy cost is a crucial requirement for unmanned and battery operated autonomous systems. Robotic applications require real time depth estimation for navigation and decision making under rapidly changing 3D surroundings. A high accuracy algorithm may provide the best depth estimation but may consume tremendous compute and energy resources. A general trade-off is to choose less accurate methods for initial depth estimate and a more accurate yet compute intensive method when needed. Previous work has shown this trade-off can be improved by developing a state-of-the-art method (AnyNet) to improve stereo depth estimation. We studied both the monocular and stereo vision depth estimation methods and investigated methods to reduce computational complexity of these methods. This was our baseline. Consequently, our experiments show reduction of monocular depth estimation model size by ~75% reduces accuracy by less than 2% (SSIM metric). Our experiments with the novel stereo vision method (AnyNet) show that accuracy of depth estimation does not degrade more than 3% (three pixel error metric) in spite of reduction in model size by ~20%. We have shown that smaller models can indeed perform competitively.