Paper Info
Title
Occlusion-robust 3D sensing using aerial imaging
Abstract
Conventional active 3D sensing systems do not work well when other objects get between the measurement target and the measurement equipment, occluding the line of sight. In this paper, we propose an active 3D sensing method that solves this occlusion problem by using a light field created using aerial imaging. In this light field, aerial luminous spots can be formed by focusing rays of light from multiple directions. Towards the occlusion problem, this configuration is effective, because even if some of the rays are occluded, the rays of other directions keep the spots. Our results showed that this method was able to measure the position and inclination of a target by using an aerial image of a single point light source and was robust against occlusions. In addition, we confirmed that multiple point light sources also worked well.
Year
DOI
Venue
2016
10.1109/ICCPHOT.2016.7492883
2016 IEEE International Conference on Computational Photography (ICCP)
Field
DocType
ISSN
Computer vision,Occlusion,Computer science,Optics,3d sensing,Aerial image,Light field,Artificial intelligence,Line-of-sight,Light source
Conference
2164-9774
Citations 
PageRank 
References 
0
0.34
16
Authors
3
Name
Order
Citations
PageRank
Masato Yasui131.42
Yoshihiro Watanabe25010.74
Masatoshi Ishikawa3903163.13