A hybrid of local and global atmospheric scattering model for depth prediction via cross Bayesian model. (28th June 2022)

• Record Type: Journal Article
• Authors: Zhao, Qianjin; Zhang, Haitao; Cui, Jianhua; Sun, Yanguang; Duan, Songsong; Xia, Chenxing; Gao, Xiuju
• Is Part Of: International journal of computational science and engineering. Volume 25:Number 4(2022)
• Page Start: 448
• Page End: 459
• Publication Date: 2022-06-28
• Subjects: depth estimation -- global atmospheric light -- local atmospheric light -- cross Bayesian model; Computer science -- Mathematics -- Periodicals; Computer simulation -- Mathematical aspects -- Periodicals; Computational intelligence -- Periodicals; 004.015105
• Languages: English
• ISSNs: 1742-7185

Abstract

Monocular depth estimation is a fascinating and challenging problem in virtual vision. However, the training of networks based on deep learning largely depends on the training data. This paper proposes a depth prediction method based on the depth cue: atmospheric light scattering, which can effectively predict the depth in different atmospheric light scenarios. But the assumption of global atmospheric light constancy can produce the unavoidable error. Especially for complex scenes, the complex reflected light of the scene leads to uneven distribution of atmospheric light. This paper proposes a new local atmospheric light estimation method, which can simulate the real distribution of atmospheric light scattering in the air more effectively. And the experiments found that the two models are complementary. In order to fuse the intrinsic real information of the two models, this paper adopts the fusion strategy based on the Bayesian model, and edge-preserving filtering is used to preserve the detailed information.