UABDivulga - Barcelona recerca i innovació

Continuous innovations in automotive lighting technology pose the problem of how to assess new headlights systems. For car manufacturers, assessment is mostly relative: given a headlights system to be tested, how does it compare to another one, maybe from a different supplier, in terms of features like light intensity, homogeneity or reach? This comparison is best performed dynamically, asking experts to actually drive along a certain testing track to later write down the visual impressions they remember.

However, this procedure suffers from several drawbacks: comparisons can not be repeated, are not retrospective and can not be either properly shared with other people since the only record is a paper form. To overcome them, we propose to record, for each headlights system, a video sequence of what the driver sees with a camera attached to the windshield screen. The problem becomes now how to compare a pair of such sequences. Two issues must be addressed: the temporal alignment or synchronization of the two sequences, and then the spatial alignment or registration of all the corresponding frames.

In this work we propose a semi-automatic but fast procedure for the former, and an automatic method for the later. In addition, we explore an alternative to the joint visualization of corresponding frames called the bird's view transform, and propose a simple fusion technique for a better visualization of the headlights differences in
two sequences. Results are provided for a number of headlights with different light sources and from several vehicle brands.

A series of still images can not convey the dynamic aspect of the results for this kind of application. Therefore, we have built a web page where a number of original and synchronized videos can be played. Please visit http://www.cvc.uab.es/adas/projects/sincro/JAE/ for the proper visualization of results.