Vehicle object retargeting from dynamic traffic videos for real-time visualisation

Journal article

One form of video visualisation is to transform traffic videos from a street view to an aerial view, which facilitates a summary overview of multiple traffic video streams. This paper presents an efficient and effective solution to mitigate the undesirable distortion of the re-targeted vehicle objects in traffic video visualisation. This is achieved by a series of automated algorithmic steps, including vehicle segmentation, vehicle roof detection, and non-uniform image deformation by applying a second homography. This technique has been integrated into a video visualisation system that creates an aerial view of re-targeted video streams on top of a conventional aerial view. The results have shown that the technique offers the system a significant improvement in visual quality without undermining the requirement for real-time video visualisation. © 2013 Springer-Verlag Berlin Heidelberg.

Authors: Walton, S; Berger, K; Ebert, D; Chen, M

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: Springer Berlin Heidelberg
• Journal: Visual Computer
• Volume: 30
• Issue: 5
• Pages: 493-505
• Publication date: 2014-05-01
• DOI: 10.1007/s00371-013-0874-5
• EISSN: 1432-2315
• ISSN: 0178-2789
• Keywords: Optical flow and edge detection; Video visualisation; Traffic visualisation; Retargeting; Car roof detection