Jan Prokaj, Ph.D.

Using 3D Scene Structure to Improve Tracking

Problem

  • Aerial imagery collected over urban areas contains many occlusions for vehicles
  • Current tracking algorithms cannot track vehicles through such occlusions due to weak object appearance and complexity of motion prediction over long period

Approach

Flowchart describing the proposed approach.
  • Use known 3D scene structure to estimate a dynamic occlusion map: binary map, which indicates what regions of the image are occluders of moving objects; dynamic due to camera motion
  • Cluster track initiations and terminations near an occlusion map to find sources and sinks
  • Find corresponding source-sink pairs using the Hungarian algorithm
  • Match a sequence of tracks at the sink with a sequence of tracks at the source using dynamic programming
  • Merge the matched tracks

Results

Proposed Algorithm Hungarian Algorithm
Correctly Linked 58% 42%
Incorrectly Linked 3.4% 5.8%

Data

Code

Reference

J. Prokaj , G. Medioni . Using 3D Scene Structure to Improve Tracking. In IEEE Conference on Computer Vision and Pattern Recognition , pages 1337-1344, 2011.

[PDF] [DOI] [Poster] [Video]

Related Publications

  • J. Prokaj . Exploitation of Wide Area Motion Imagery . Ph.D. thesis, University of Southern California, 2013.

    [PDF]