New features and insights for pedestrian detection

• Published in: 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition pp. 1030 - 1037
• Main Authors: Walk, Stefan, Majer, Nikodem, Schindler, Konrad, Schiele, Bernt
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.06.2010
• Subjects: Cascading style sheets; Detectors; Feature extraction; Histograms; Humans; Image motion analysis; Image sequences; Object detection; Optical saturation; Optical sensors
• ISBN: 1424469848; 9781424469840
• ISSN: 1063-6919
• DOI: 10.1109/CVPR.2010.5540102

Despite impressive progress in people detection the performance on challenging datasets like Caltech Pedestrians or TUD-Brussels is still unsatisfactory. In this work we show that motion features derived from optic flow yield substantial improvements on image sequences, if implemented correctly - even in the case of low-quality video and consequently degraded flow fields. Furthermore, we introduce a new feature, self-similarity on color channels, which consistently improves detection performance both for static images and for video sequences, across different datasets. In combination with HOG, these two features outperform the state-of-the-art by up to 20%. Finally, we report two insights concerning detector evaluations, which apply to classifier-based object detection in general. First, we show that a commonly under-estimated detail of training, the number of bootstrapping rounds, has a drastic influence on the relative (and absolute) performance of different feature/classifier combinations. Second, we discuss important intricacies of detector evaluation and show that current benchmarking protocols lack crucial details, which can distort evaluations.