Multiple Prior Representation Learning for Self-Supervised Monocular Depth Estimation via Hybrid Transformer

TL;DR

This paper presents a self-supervised monocular depth estimation model that integrates multiple priors, including spatial, context, and semantic information, using a hybrid transformer to improve depth prediction accuracy in complex scenes.

Contribution

The paper introduces a novel multi-prior approach utilizing a hybrid transformer and semantic boundary loss to enhance scene understanding in depth estimation.

Findings

Improved depth estimation accuracy across three datasets.

Enhanced generalization in complex and untextured regions.

Effective integration of spatial, context, and semantic priors.

Abstract

Self-supervised monocular depth estimation aims to infer depth information without relying on labeled data. However, the lack of labeled information poses a significant challenge to the model's representation, limiting its ability to capture the intricate details of the scene accurately. Prior information can potentially mitigate this issue, enhancing the model's understanding of scene structure and texture. Nevertheless, solely relying on a single type of prior information often falls short when dealing with complex scenes, necessitating improvements in generalization performance. To address these challenges, we introduce a novel self-supervised monocular depth estimation model that leverages multiple priors to bolster representation capabilities across spatial, context, and semantic dimensions. Specifically, we employ a hybrid transformer and a lightweight pose network to obtain long-range spatial priors in the spatial dimension. Then, the context prior attention is designed to improve generalization, particularly in complex structures or untextured areas. In addition, semantic priors are introduced by leveraging semantic boundary loss, and semantic prior attention is supplemented, further refining the semantic features extracted by the decoder. Experiments on three diverse datasets demonstrate the effectiveness of the proposed model. It integrates multiple priors to comprehensively enhance the representation ability, improving the accuracy and reliability of depth estimation. Codes are available at: \url{https://github.com/MVME-HBUT/MPRLNet}