Adding Model Constraints to CNN for Top View Hand Pose Recognition in Range Images

Adding Model Constraints to CNN for Top View Hand Pose Recognition in Range Images
Aditya Tewari, Frederic Grandidier, Bertram Taetz, Didier Stricker
Proceedings of the 5th International Conference in Pattern Recognition Applications and Methods ICPRAM 2016 International Conference on Pattern Recognition Applications and Methods (ICPRAM-05), 5th, February 24-26, Rome, Italy

Abstract:
A new dataset for hand-pose is introduced. The dataset includes the top view images of the palm by Time of Flight (ToF) camera. It is recorded in an experimental setting with twelve participants for six hand-poses. An evaluation on the dataset is carried out with a dedicated Convolutional Neural Network (CNN) architecture for Hand Pose Recognition (HPR). This architecture uses a model-layer. The small size model layer creates a funnel shape network which adds a priori knowledge and constrains the network by modelling the degree of freedom of the palm, such that it learns palm features. It is demonstrated that this network performs better than a similar network without the prior added. A two-phase learning scheme which allows training the model on full dataset even when the classification problem is confined to a subset of the classes is described. The best model performs at an accuracy of 92%. Finally, we show the feature transfer capability of the network and compare the extracted features from various networks and discuss usefulness for various applications.
Keywords:
CNN , Hand-Pose, Feature Transfer, Transfer Learning, Fine Tuning.