Frequency-Aware Self-Supervised Monocular Depth Estimation

TL;DR

This paper introduces frequency-aware techniques, including ambiguity-masking and a frequency-adaptive Gaussian filter, to enhance self-supervised monocular depth estimation by addressing photometric loss issues at object boundaries and high-frequency regions.

Contribution

The paper proposes novel, lightweight methods that improve existing depth estimation models by solving fundamental photometric loss problems without increasing inference complexity.

Findings

Performance improvements on multiple models including state-of-the-art.

No additional parameters or inference cost introduced.

Effective handling of boundary ambiguities and high-frequency regions.

Abstract

We present two versatile methods to generally enhance self-supervised monocular depth estimation (MDE) models. The high generalizability of our methods is achieved by solving the fundamental and ubiquitous problems in photometric loss function. In particular, from the perspective of spatial frequency, we first propose Ambiguity-Masking to suppress the incorrect supervision under photometric loss at specific object boundaries, the cause of which could be traced to pixel-level ambiguity. Second, we present a novel frequency-adaptive Gaussian low-pass filter, designed to robustify the photometric loss in high-frequency regions. We are the first to propose blurring images to improve depth estimators with an interpretable analysis. Both modules are lightweight, adding no parameters and no need to manually change the network structures. Experiments show that our methods provide performance boosts to a large number of existing models, including those who claimed state-of-the-art, while introducing no extra inference computation at all.