Training Self-Supervised Depth Completion Using Sparse Measurements and a Single Image

TL;DR

This paper introduces a self-supervised depth completion method that uses only sparse depth measurements and a single image, eliminating the need for dense labels or multiple viewpoints, and employs novel loss functions and segmentation maps.

Contribution

It proposes a new self-supervised approach for depth completion that works with only sparse measurements and a single image, avoiding dense labels and multi-view data.

Findings

Effective depth propagation from sparse points to unobserved regions.

Improved depth estimation accuracy demonstrated through extensive experiments.

Segmentation maps enhance depth prediction quality.

Abstract

Depth completion is an important vision task, and many efforts have been made to enhance the quality of depth maps from sparse depth measurements. Despite significant advances, training these models to recover dense depth from sparse measurements remains a challenging problem. Supervised learning methods rely on dense depth labels to predict unobserved regions, while self-supervised approaches require image sequences to enforce geometric constraints and photometric consistency between frames. However, acquiring dense annotations is costly, and multi-frame dependencies limit the applicability of self-supervised methods in static or single-frame scenarios. To address these challenges, we propose a novel self-supervised depth completion paradigm that requires only sparse depth measurements and their corresponding image for training. Unlike existing methods, our approach eliminates the need for dense depth labels or additional images captured from neighboring viewpoints. By leveraging the characteristics of depth distribution, we design novel loss functions that effectively propagate depth information from observed points to unobserved regions. Additionally, we incorporate segmentation maps generated by vision foundation models to further enhance depth estimation. Extensive experiments demonstrate the effectiveness of our proposed method.