Three-dimensional reconstruction of heavily occluded pedestrians using integral imaging

Ghaneizad, M ; Sharif University of Technology

368 Viewed
  1. Type of Document: Article
  2. DOI: 10.1145/2967413.2967417
  3. Publisher: Association for Computing Machinery
  4. Abstract:
  5. In this paper, we propose a novel approach for handling the occlusion problem in pedestrian detection through optical principles. Our proposed framework is based on a popular three-dimensional imaging technique in optics, named integral imaging, in which multiple viewpoint images captured from a three-dimensional scene are used to extract range information of the scene. The proposed approach effectively reconstructs an unobstructed view of heavily occluded pedestrians simultaneously containing range information. The range information provided by our method can be used for foreground extraction and the reconstruction output can be used in various applications, such as people detection and tracking. The validity of our novel approach has been demonstrated by experimental results for different scenarios
  6. Keywords:
  7. Cameras ; Imaging systems ; Imaging techniques ; Object detection ; Target tracking ; Tracking (position) ; Foreground extraction ; Heavy occlusion ; Integral imaging ; Multiple viewpoints ; Pedestrian detection ; Three dimensional imaging ; Three-dimensional reconstruction ; Three-dimensional scenes ; Image processing
  8. Source: 10th International Conference on Distributed Smart Cameras, ICDSC 2016, 12 September 2016 through 15 September 2016 ; Volume 12-15-September-2016 , 2016 , Pages 1-7 ; 9781450347860 (ISBN)
  9. URL: http://dl.acm.org/citation.cfm?id=2967417