English

FusionDepth: Complement Self-Supervised Monocular Depth Estimation with Cost Volume

Computer Vision and Pattern Recognition 2023-05-11 v1

Abstract

Multi-view stereo depth estimation based on cost volume usually works better than self-supervised monocular depth estimation except for moving objects and low-textured surfaces. So in this paper, we propose a multi-frame depth estimation framework which monocular depth can be refined continuously by multi-frame sequential constraints, leveraging a Bayesian fusion layer within several iterations. Both monocular and multi-view networks can be trained with no depth supervision. Our method also enhances the interpretability when combining monocular estimation with multi-view cost volume. Detailed experiments show that our method surpasses state-of-the-art unsupervised methods utilizing single or multiple frames at test time on KITTI benchmark.

Keywords

Cite

@article{arxiv.2305.06036,
  title  = {FusionDepth: Complement Self-Supervised Monocular Depth Estimation with Cost Volume},
  author = {Zhuofei Huang and Jianlin Liu and Shang Xu and Ying Chen and Yong Liu},
  journal= {arXiv preprint arXiv:2305.06036},
  year   = {2023}
}
R2 v1 2026-06-28T10:30:54.405Z