The LaRED is recorded with the Intel’s short range depth camera (the model number: VF0780-SDK-CE). In particular, the LaRED dataset contains 27 gestures in 3 different orientations, which makes a total of 81 classes. We have 10 subjects performing the corresponding hand gesture and 300 images are collected from each subject with the Intel camera providing a pair of synchornized color and depth images. Meanwhile, a mask image is given to localize the boundary of the gesture-performing hand.
Data and code available in archives: Link: http://mclab.citi.sinica.edu.tw/dataset/lared/lared.html
References and Citation
Use of the datasets in published work should be acknowledged by a full citation to the authors' papers [HSL14] at the MMSys conference (Proceedings of ACM MMSys 2014, March 19 - March 21, 2014, Singapore, Singapore).
HSL14: Yuan-Sheng Hsiao, Jordi Sanchez-Riera, Tekoing Lim, Kai-Lung Hua, Wen-Huang Cheng, LaRED: a large RGB-D extensible hand gesture dataset, Proceedings of ACM MMSys 2014, March 19 - March 21, 2014, Singapore, Singapore.