WTA Hash-Based Multimodal Feature Fusion for 3D Human Action Recognition | IEEE Conference Publication | IEEE Xplore