| title | Learning to Discriminate Scene Structures Makes Self-Supervised Depth Learning Scalable |
|---|---|
| authors | Mengtan Zhang, Yuchuan Cui, Yu Ma, Zizhan Guo, Wei Ye, Rui Fan |
| date | 2026-10-16 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2610.23600 |
| affiliation | Tongji Unversity |
| upvotes | 7 (Hugging Face Daily Papers, 2026-11-01) |
| links | abs pdf hf code |
| abstract | Self-supervised monocular depth estimation is appealing for its potential to scale with unlabeled data, yet existing methods often degrade when trained on diverse scenes. This study finds that, under self-supervision, models tend to encode dataset-specific structural biases rather than transferable scene geometry, leading to cross-domain optimization conflicts. To address this issue, this study proposes the MaSS framework, aiming to make self-supervised depth learning scalable to diverse data. MaSS introduces a dedicated structural representation pathway explicitly decoupled from contextual features, which is guided by a novel scene-structure discrimination objective to learn generalized, layout-aware structural representations for depth decoding. Trained and evaluated on four diverse datasets, both individually and jointly, MaSS substantially outperforms prior self-supervised methods and is the first to consistently benefit from mixed-domain training. Zero-shot evaluation on eight additional datasets further demonstrates its remarkable generalizability, particularly on those entirely out-of-distribution scenes. The source code is publicly available at https://mias.group/MaSS. |
cite
@misc{zhang2026learning,
title={Learning to Discriminate Scene Structures Makes Self-Supervised Depth Learning Scalable},
author={Mengtan Zhang and Yuchuan Cui and Yu Ma and Zizhan Guo and Wei Ye and Rui Fan},
year={2026},
eprint={2610.23600},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2610.23600}
}