Towards Cross-View-Consistent Self-Supervised Surround Depth Estimation

TL;DR

This paper introduces a cross-view consistency approach with novel loss functions and pose estimation techniques to improve self-supervised surround depth estimation, achieving state-of-the-art results on autonomous driving datasets.

Contribution

It proposes new loss functions and a pose estimation method that explicitly enforce cross-view consistency in self-supervised surround depth estimation.

Findings

Achieves state-of-the-art performance on DDAD and nuScenes datasets.

Introduces a dense depth consistency loss and multi-view reconstruction consistency loss.

Demonstrates the effectiveness of the proposed techniques across different models.

Abstract

Depth estimation is a cornerstone for autonomous driving, yet acquiring per-pixel depth ground truth for supervised learning is challenging. Self-Supervised Surround Depth Estimation (SSSDE) from consecutive images offers an economical alternative. While previous SSSDE methods have proposed different mechanisms to fuse information across images, few of them explicitly consider the cross-view constraints, leading to inferior performance, particularly in overlapping regions. This paper proposes an efficient and consistent pose estimation design and two loss functions to enhance cross-view consistency for SSSDE. For pose estimation, we propose to use only front-view images to reduce training memory and sustain pose estimation consistency. The first loss function is the dense depth consistency loss, which penalizes the difference between predicted depths in overlapping regions. The second one is the multi-view reconstruction consistency loss, which aims to maintain consistency between reconstruction from spatial and spatial-temporal contexts. Additionally, we introduce a novel flipping augmentation to improve the performance further. Our techniques enable a simple neural model to achieve state-of-the-art performance on the DDAD and nuScenes datasets. Last but not least, our proposed techniques can be easily applied to other methods. The code is available at https://github.com/denyingmxd/CVCDepth.