Forecasting human motion from a sequence of 3D poses is an important problem in the fields of computer vision and robotics. Sequence-to-sequence learning approaches have been applied to human motion prediction, such as Recurrent Neural Networks (RNNs), Graph Convolutional Networks (GCNs) and Transformer.
This project is for RNN based methods for human motion prediction, especially the full code of our two papers [1, 2].
[1] Wang, Hongsong, et al. "PVRED: A Position-Velocity Recurrent Encoder-Decoder for Human Motion Prediction." IEEE Transactions on Image Processing 30 (2021): 6096-6106.
https://arxiv.org/pdf/1906.06514.pdf
[2] Wang, Hongsong, et al. "Velocity-to-Velocity Human Motion Forecasting." Pattern Recognition. https://www.researchgate.net/publication/356058006_Velocity-to-Velocity_Human_Motion_Forecasting