Low-Cost 3D Model Acquisition for Rapid Accident Investigation

Vehicular accidents are one of the great societal challenges. In some demographics, they are the leading cause of death. It is important that accidents are thoroughly investigated. The immediate reason is to find out who is responsible and liable for the damages. But just as important is it to collect information about accidents to determine if changes to vehicles, infrastructure or policy could prevent or mitigate future accidents. One set of tools in the investigation uses 3D models of the scene. They are usually acquired with laser scanners. However, they are very costly in time and money and are therefore only used in severe cases, like fatal accidents. In recent years digital cameras and computer vision algorithms have become so inexpensive, powerful and efficient that it is now possible to create 3D models from a set of digital images at a very low cost. In previous research the authors have shown that one can do this for vehicle accidents and that these 3D models are useful for accident investigation. In this project the authors worked on shape completing, parts segmentation, unsupervised multi-view stereopsis, and 3D transformer networks. The long-term goal is to have a set of tools that can automatically analyze an accident from a set of images. In this report the authors describe two steps towards that goal: Point cloud completion and parts separation.

Language

  • English

Media Info

  • Media Type: Digital/other
  • Edition: Final Research Report
  • Features: Figures; Photos; References;
  • Pagination: 5p

Subject/Index Terms

Filing Info

  • Accession Number: 01730541
  • Record Type: Publication
  • Contract Numbers: 69A3551747111
  • Files: UTC, TRIS, ATRI, USDOT
  • Created Date: Feb 5 2020 3:20PM