EXIT

We retrain the model without text condition for this evaluation.

Correct. Only text conditions change (for both training and testing) everything else remains the same.

Hi, You can download the [pretrain KIT](https://drive.google.com/file/d/1aZTEcZQgryx4Y1Gy40anT4K0TVRJhlnZ/view?usp=drive_link). I haven't tested it, so there may be some bugs in KIT. You can investigate the saved source code in ```./source``` P.S. For...

Hi, The default parameters are the parameters we used for training. Except the batch size, we use 512 since we use 4 gpus for training therefore we decrease `total-iter`, `lr-scheduler`,...

Thank you for your interest. We plan to release the code next month.

Hi, Thank you for your interest. We don't have the evaluation code using test set for VQ. However, you can easily modify the [train_vq](https://github.com/exitudio/MMM/blob/main/train_vq) code. 1. Change the evaluation from...

Hi, I will add more detail about rendering function soon. Meanwhile, you can follow instruction from [MDM](https://github.com/GuyTevet/motion-diffusion-model?tab=readme-ov-file#render-smpl-mesh)

Hi, We have created sample codes and visualizations in the ```edit.ipynb notebook```. Please refer to the 'Long Range' section in ```edit.ipynb```. The model requires a list of text and the...

Thank you for your interest in our work. We haven't cleaned up the upper body training code yet, but if you'd like, you can email me, and I'll send you...

Hi, You can get the code [here](https://drive.google.com/file/d/1LZwdnnKzHYa-W6mKzwVuTfmYKyJnncSR/view?usp=drive_link) For training upper-lower: 1. train VQVAE: - see shell script ./experiments/train_vq.sh - add "--sep-uplow" option to the train_vq.py 2. train transformer: - see...