Accurate Face Detection for Privacy in Street-view Mapping Applications Combining Face Boosting, Body Boosting and Skin Tone Detectors
Résumé
In the last two years, web-based applications us- ing street-level images have been developing fast. In that context, privacy preservation is an unavoidable issue. We present in this paper a multi-boosting based approach to detect pedestrians in high resolution panoramics in order to blur their faces. This task is quite complex since these features vary in size, shape, color, and often are partially occluded, sometimes be- hind windows or inside cars, etc. Our strategy is thus based on the combination of two existing boosting algo- rithms detecting faces [1] and bodies [2] with a skin tone detection algorithm we developed. The results are quite encouraging for such an unconstrained data: 86:2% of true positives and an average of 2 false positive de- tections per image (2.1 MPixels). This combination solution provides much more robust results than each detection algorithm performed independently.