Monocular catadioptric panoramic depth estimation via improved end-to-end neural network model

doi:10.3389/fnbot.2023.1278986

CIOMP OpenIR

	Monocular catadioptric panoramic depth estimation via improved end-to-end neural network model
	F. Yan, L. Liu, X. P. Ding, Q. Zhang and Y. Q. Liu
	2023
发表期刊	Frontiers in Neurorobotics
ISSN	1662-5218
卷号	17 期号:11
摘要	In this paper, we propose a monocular catadioptric panoramic depth estimation algorithm based on an improved end-to-end neural network model. First, we use an enhanced concentric circle approximation unfolding algorithm to unfold the panoramic images captured by the catadioptric panoramic camera and then extract the effective regions. In addition, the integration of the Non-local attention mechanism is exploited to improve image understanding. Finally, a depth smoothness loss strategy is implemented to further enhance the reliability and precision of the estimated depths. Experimental results confirm that this refined algorithm is capable of providing highly accurate estimates of object depth.
DOI	10.3389/fnbot.2023.1278986
URL	查看原文
收录类别	sci
语种	英语
引用统计
文献类型	期刊论文
条目标识符	http://ir.ciomp.ac.cn/handle/181722/68072
专题	中国科学院长春光学精密机械与物理研究所
推荐引用方式 GB/T 7714	F. Yan, L. Liu, X. P. Ding, Q. Zhang and Y. Q. Liu. Monocular catadioptric panoramic depth estimation via improved end-to-end neural network model[J]. Frontiers in Neurorobotics,2023,17(11).
APA	F. Yan, L. Liu, X. P. Ding, Q. Zhang and Y. Q. Liu.(2023).Monocular catadioptric panoramic depth estimation via improved end-to-end neural network model.Frontiers in Neurorobotics,17(11).
MLA	F. Yan, L. Liu, X. P. Ding, Q. Zhang and Y. Q. Liu."Monocular catadioptric panoramic depth estimation via improved end-to-end neural network model".Frontiers in Neurorobotics 17.11(2023).

条目包含的文件		下载所有文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
Monocular catadioptr（2028KB）	期刊论文	出版稿	开放获取	CC BY-NC-SA	浏览下载