Multimodal Variational Autoencoders for Sensor Fusion and Cross Generation | IEEE Conference Publication | IEEE Xplore