Wei Mao

It could because of your pytorch version. Please find our version in readme file.

Hi, Thank you for your interests. We do not have the preprocess code from raw h3.6 data to expmap either. We obtain the data from the structure-rnn paper. https://github.com/asheshjain399/RNNexp/tree/srnn/structural_rnn

Hi. Thank you for your interests. Those parameters are actually in the readme file where I provide the running commands for each datasets.

Hi, Thank you for your interests. It is common to be fixed for long-term prediction (>25 frames). It is currently an open question in this field. I do not think...

Hi, you need set num_seed to a number greater than 1. In our results we set it to be 5. It means using different number of random seed. The code...

Thank you for your interests. The forward function of body prior model will first compute the mesh vertex then from those vertex calculate the 3D joints. They are too costly....

Thank you for your interests. There are many ways of visualizing then, you can refer to [here](https://github.com/una-dinosauria/human-motion-prediction ) for details.

Thank you for your interests. One for prediction in "3D" space, one for angle space.