EC-Depth: Exploring the consistency of self-supervised monocular depth estimation in challenging scenes
Preprint
Abstract
Self-supervised monocular depth estimation holds significant importance in the fields of autonomous driving and robotics. However, existing methods are typically trained and tested on standard datasets, overlooking the impact of various adverse conditions prevalent in real-world applications, such as rainy days. As a result, it is commonly observed that these methods struggle to handle these challenging scenarios. To address this issue, we present EC-Depth, a novel self-supervised two-stage framework to achieve a robust depth estimation. In the first stage, we propose depth consistency regularization to propagate reliable supervision from standard to challenging scenes. In the second stage, we adopt the Mean Teacher paradigm and propose a novel consistency-based pseudo-label filtering strategy to improve the quality of pseudo-labels, further improving both the accuracy and robustness of our model. Extensive experiments demonstrate that our method achieves accurate and consistent depth predictions in both standard and challenging scenarios, surpassing existing state-of-the-art methods on KITTI, KITTI-C, DrivingStereo, and NuScenes-Night benchmarks.Keywords
All Related Versions
This publication has 11 references indexed in Scilit:
- Self-supervised Monocular Depth Estimation: Let’s Talk About The Weather2023
- Robust Monocular Depth Estimation under Challenging Conditions2023
- MonoViT: Self-Supervised Monocular Depth Estimation with a Vision Transformer2022
- Hierarchical Object Relationship Constrained Monocular Depth Estimation.Pattern Recognition, 2021
- EnlightenGAN: Deep Light Enhancement Without Paired SupervisionIEEE Transactions on Image Processing, 2021
- Mobile3DRecon: Real-time Monocular 3D Reconstruction on a Mobile PhoneIEEE Transactions on Visualization and Computer Graphics, 2020
- Rain Rendering for Evaluating and Improving Robustness to Bad WeatherInternational Journal of Computer Vision, 2020
- Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-Shot Cross-Dataset TransferIEEE Transactions on Pattern Analysis and Machine Intelligence, 2020
- Consistent video depth estimationACM Transactions on Graphics, 2020
- Image quality assessment: from error visibility to structural similarityIEEE Transactions on Image Processing, 2004