If you use this code in any context, please cite the following paper:
@misc{oreshkin2022motion,
title={Motion Inbetweening via Deep $\Delta$-Interpolator},
author={Boris N. Oreshkin and Antonios Valkanas and Félix G. Harvey and Louis-Simon Ménard and Florent Bocquelet and Mark J. Coates},
year={2022},
eprint={2201.06701},
archivePrefix={arXiv},
primaryClass={cs.LG}
}
mkdir workspace
cd workspace
git clone https://github.com/boreshkinai/delta-interpolator
Build image and start the lightweight docker container. Note that this assumes that the data for the project will be stored in the shared folder /home/pose-estimation accessible to you and other project members.
docker build -f Dockerfile -t delta_interpolator:$USER .
nvidia-docker run -p 18888:8888 -p 16006:6006 -v ~/workspace/delta-interpolator:/workspace/delta-interpolator -t -d --shm-size="8g" --name delta_interpolator_$USER delta_interpolator:$USER
docker exec -i -t delta_interpolator_$USER /bin/bash
Once inside docker container, this launches the training session for the proposed model. Checkpoints and tensorboard logs are stored in ./logs/lafan/transformer
python run.py --config=src/configs/transformer.yaml
This evaluates zero-velocity and the interpolator models
python run.py --config=src/configs/interpolator.yaml
python run.py --config=src/configs/zerovel.yaml
Training losses eveolve as follows:
http://your_server_ip:18888/notebooks/LaFAN1Results.ipynb
The notebook password is default
https://storage.googleapis.com/delta-interpolator/pretrained_model.zip
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent