Real-time fingertip localization conditioned on hand gesture classification

A method to obtain accurate hand gesture classification and fingertip localization from depth images is proposed. The Oriented Radial Distribution feature is utilized, exploiting its ability to globally describe hand poses, but also to locally detect likely fingertip positions. Hence, hand gesture and fingertip locations are characterized with a single feature calculation. We propose to divide the difficult problem of locating fingertips into two more tractable problems, taking advantage of hand gesture as an auxiliary variable. Besides, the ColorTip dataset is proposed, a dataset for hand gesture recognition and fingertip classification on depth data. ColorTip allows automatic fingertip annotation through a wieldy and not costly footage. The proposed method is evaluated against recent works and datasets, achieving promising results in both gesture classification and fingertip localization.