Title :
Analysis of rotational robustness of hand detection with a Viola-Jones detector
Author :
Kölsch, Mathias ; Turk, Matthew
Author_Institution :
Dept. of Comput. Sci., California Univ., Santa Barbara, CA, USA
Abstract :
The research described in this paper analyzes the in-plane rotational robustness of the Viola-Jones object detection method when used for hand appearance detection. We determine the rotational bounds for training and detection for achieving undiminished performance without an increase in classifier complexity. The result - up to 15° total - differs from the method´s performance on faces (30° total). We found that randomly rotating the training data within these bounds allows for detection rates about one order of magnitude better than those trained on strictly aligned data. The implications of the results effect both savings in training costs as well as increased naturalness and comfort of vision-based hand gesture interfaces.
Keywords :
computer vision; gesture recognition; image classification; learning (artificial intelligence); object detection; set theory; Viola-Jones detector; hand appearance detection; image classification; inplane rotational robustness analysis; learning based object detection method; set theory; vision based hand gesture interfaces; Cameras; Computer science; Costs; Detectors; Face detection; Object detection; Pattern recognition; Robustness; Skin; Training data;
Conference_Titel :
Pattern Recognition, 2004. ICPR 2004. Proceedings of the 17th International Conference on
Print_ISBN :
0-7695-2128-2
DOI :
10.1109/ICPR.2004.1334480