High-Precision Self-Supervised Monocular Depth Estimation with Rich-Resource Prior

TL;DR

This paper introduces RPrDepth, a self-supervised monocular depth estimation method that leverages rich-resource prior information during training to achieve high accuracy with only single-image input at inference.

Contribution

The proposed RPrDepth method uses offline extracted rich-resource features as priors, enabling accurate depth estimation from a single image without requiring rich-resource inputs during inference.

Findings

Outperforms other single-image depth models.

Achieves comparable or better results than rich-resource input models.

Uses offline prior features to enhance single-image depth estimation.

Abstract

In the area of self-supervised monocular depth estimation, models that utilize rich-resource inputs, such as high-resolution and multi-frame inputs, typically achieve better performance than models that use ordinary single image input. However, these rich-resource inputs may not always be available, limiting the applicability of these methods in general scenarios. In this paper, we propose Rich-resource Prior Depth estimator (RPrDepth), which only requires single input image during the inference phase but can still produce highly accurate depth estimations comparable to rich resource based methods. Specifically, we treat rich-resource data as prior information and extract features from it as reference features in an offline manner. When estimating the depth for a single-image image, we search for similar pixels from the rich-resource features and use them as prior information to estimate the depth. Experimental results demonstrate that our model outperform other single-image model and can achieve comparable or even better performance than models with rich-resource inputs, only using low-resolution single-image input.