GGTr: An Innovative Framework for Accurate and Realistic Human Motion Prediction

Human motion prediction involves forecasting future movements based on past observations, which is a complex task due to the inherent spatial-temporal dynamics of human motion. In this paper, we introduced novel framework, GGTr, adeptly encapsulates these patterns by integrating positional graph convolutional network (GCN) layers, gated recurrent unit (GRU) and transformer layers. The proposed model utilizes an enhanced GCN layer equipped with representation aggregate information from body joints more effectively. To address temporal dependencies, strategically combined GRU enabling capture both local global dependencies across joints. Through extensive experiments conducted Human3.6M CMU-MoCap datasets, demonstrated superior performance our model. Notably, framework shows significant improvements in predicting long-term movements, outperforming state-of-the-art methods substantially.