Space-Time Domain Tensor Neural Networks: An Application on Human Pose Classification
This work addresses efficient pattern recognition for spatiotemporal data in applications like human pose analysis, but it appears incremental as it builds on existing tensor and neural network methods.
The paper tackles human pose classification from 3D skeleton data by proposing a tensor-based neural network with novel spatiotemporal feature construction, fusion, and processing, achieving state-of-the-art performance.
Recent advances in sensing technologies require the design and development of pattern recognition models capable of processing spatiotemporal data efficiently. In this study, we propose a spatially and temporally aware tensor-based neural network for human pose classification using three-dimensional skeleton data. Our model employs three novel components. First, an input layer capable of constructing highly discriminative spatiotemporal features. Second, a tensor fusion operation that produces compact yet rich representations of the data, and third, a tensor-based neural network that processes data representations in their original tensor form. Our model is end-to-end trainable and characterized by a small number of trainable parameters making it suitable for problems where the annotated data is limited. Experimental evaluation of the proposed model indicates that it can achieve state-of-the-art performance.