Journal
IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE
Volume 6, Issue 5, Pages 1237-1246Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TETCI.2022.3182360
Keywords
Estimation; Training; Adaptation models; Decoding; Lighting; Computational intelligence; Sun; Unsupervised estimation; domain adaptation; monocular depth estimation; night; rainy night
Categories
Funding
- National Natural Science Foundation of China Basic Science Center Program [61988101]
- National Natural Science Fund for Distinguished Young Scholars [61725301]
- Programme of Introducing Talents of Discipline to Universities the 111 Project [B17017]
- Program of Shanghai Academic Research Leader [20XD1401300]
- Innovation Research Funding of China National Petroleum Corporation [2021D002-0902]
Ask authors/readers for more resources
This paper investigates the problem of unsupervised monocular depth estimation in highly complex scenarios and addresses this challenging problem by adopting an image transfer-based domain adaptation framework. Extensive experiments show the effectiveness of the proposed unsupervised framework in estimating the dense depth map from highly complex images.
With the development of computational intelligence algorithms, unsupervised monocular depth and pose estimation framework, which is driven by warped photometric consistency, has shown great performance in the day-time scenario. While in some challenging environments, like night and rainy night, the essential photometric consistency hypothesis is untenable because of the complex lighting and reflection, so that the above unsupervised framework cannot be directly applied to these complex scenarios. In this paper, we investigate the problem of unsupervised monocular depth estimation in highly complex scenarios and address this challenging problem by adopting an image transfer-based domain adaptation framework. We adapt the depth model trained on day-time scenarios to be applicable to night-time scenarios, and constraints on both feature space and output space promote the framework to learn the key features for depth decoding. Meanwhile, we further tackle the effects of unstable image transfer quality on domain adaptation, and an image adaptation approach is proposed to evaluate the quality of transferred images and re-weight the corresponding losses, so as to improve the performance of the adapted depth model. Extensive experiments show the effectiveness of the proposed unsupervised framework in estimating the dense depth map from highly complex images.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available