Diffusion-Augmented Depth Prediction with Sparse Annotations

TL;DR

This paper introduces DADP, a supervised depth prediction framework that uses diffusion models and object-guided loss to improve depth structure accuracy and robustness in autonomous driving scenes with sparse annotations.

Contribution

The paper proposes a novel diffusion-augmented supervised framework with an object-guided integrality loss for better depth estimation under sparse annotations.

Findings

Significant improvements in depth structure accuracy.

Enhanced robustness in depth predictions.

Effective application on three driving benchmarks.

Abstract

Depth estimation aims to predict dense depth maps. In autonomous driving scenes, sparsity of annotations makes the task challenging. Supervised models produce concave objects due to insufficient structural information. They overfit to valid pixels and fail to restore spatial structures. Self-supervised methods are proposed for the problem. Their robustness is limited by pose estimation, leading to erroneous results in natural scenes. In this paper, we propose a supervised framework termed Diffusion-Augmented Depth Prediction (DADP). We leverage the structural characteristics of diffusion model to enforce depth structures of depth models in a plug-and-play manner. An object-guided integrality loss is also proposed to further enhance regional structure integrality by fetching objective information. We evaluate DADP on three driving benchmarks and achieve significant improvements in depth structures and robustness. Our work provides a new perspective on depth estimation with sparse annotations in autonomous driving scenes.