Visibility, occlusion, and the aspect graph
International Journal of Computer Vision
Make3D: Learning 3D Scene Structure from a Single Still Image
IEEE Transactions on Pattern Analysis and Machine Intelligence
| Hi-index | 0.00 |
This paper presents a unique formulation and concept of the dynamic object effect on constant objects, such as buildings, dealing with visibility problems in 3D urban environments. Dynamic objects in a 3D urban environment, such as cars, pedestrians and trees, are usually omitted from visibility analysis. In order to challenge this problem, we focus on modeling predicting and estimating dynamic objects' future location in the environment. We integrate all these factors into our visibility analysis and create a probabilistic visibility analysis, changed over time due to the dynamic character of these objects. Our probabilistic visibility concept takes into account 3D boxes and cylinders, generating a fast and exact analytic solution to dynamic and static objects; to illustrate our concept, we use web-cameras located at constant points in the treated environments in order to update our model in each time period from web source data and to analyze the environment. Dynamic objects prediction is based on validated models for driver behavior; pedestrians' walking routes, trees' displacement and wind effect. A real urban environment with dynamic objects approximated by 3D boxes and cylinders demonstrates our approach.