Skip to content

Latest commit

 

History

History
36 lines (27 loc) · 1.91 KB

README.md

File metadata and controls

36 lines (27 loc) · 1.91 KB

MonoDiffusion: Self-Supervised Monocular Depth Estimation Using Diffusion Model

Shuwei Shao1  Zhongcai Pei1Weihai Chen1  Dingchi Sun1Peter C. Y. Chen2Zhengguo Li3
1Beihang University, 2National University of Singapore, 3A*STAR

Arxiv 2023

Abstract

We introduce a novel self-supervised depth estimation framework, dubbed MonoDiffusion, by formulating it as an iterative denoising process. Because the depth ground-truth is unavailable in the training phase, we develop a pseudo ground-truth diffusion process to assist the diffusion in MonoDiffusion. The pseudo ground-truth diffusion gradually adds noise to the depth map generated by a pre-trained teacher model. Moreover, the teacher model allows applying a distillation loss to guide the denoised depth. Further, we develop a masked visual condition mechanism to enhance the denoising ability of model. Extensive experiments are conducted on the KITTI and Make3D datasets and the proposed MonoDiffusion outperforms prior state-of-the-art competitors.
The source code is comming!