Improving the Performance of Unimodal Dynamic Hand-Gesture Recognition with Multimodal Training
This research introduces a multimodal training framework for 3D-CNNs that leverages spatiotemporal alignment and focal regularization to achieve state-of-the...