The Tornado-based web server provdes a REST interface to an SQLite database for the management of skeleton animation data and machine learning models using a collections, tags and skeletons. The database is integrated with anim_utils using a Python client. To visualize the motions in the database a Unity WebGL client is provided.
To edit and annotate data in the database, the motion_preprocessing_tool can be used. Motion editing functions can also be registered as data transforms to be executed via the web client on a batch of motions.
-
Install Python 3.6 or above in a virtual environment.
-
Clone the repository with all submodules.
git clone --recursive git@github.com:eherr/motion_database_server.git
- Install the base packages for animation data editing and Python API
pip install git+https://github.com/eherr/anim_utils
pip install git+https://github.com/eherr/motion_db_interface
- Install other dependencies
pip install -r requirements.txt
- Create a new database:
python create_database.py PROJECT_NAME ADMIN_NAME ADMIN_PASSWORD ADMIN_EMAIL
- Import default data types and data transforms:
python import_data_types.py default_data_types.json
- Add a first skeleton using an example BVH file:
python import_skeleton.py SKELETON_NAME BVH_FILE
- Import BVH files from a directory specifying the previously imported skeleton:
python import_bvh_from_directory.py PROJECT_NAME SKELETON_NAME DIRECTORY_PATH
- Start the web server:
python main.py
- Build the web client using angular:
cd webclient
ng build
-
Open the URL "localhost:8888" in browser to view the using the web client. Login using the admin user, to be able to upload motions. The port can be changed in db_server_config.json.
-
To upload and edit animations or upload and edit skeletons for retargeting you can also use the motion_preprocessing_tool.
Herrmann, E., Du, H., Antalki, A., Rubinstein, D., Schubotz, R., Sprenger, J., Hosseini, S., Cheema, N., Zinnikus, I., Manns, M., Fischer, K. Slusallek, P., "Motion Data and Model Management for Applied Statistical Motion Synthesis" In: Proceedings of the Conference on Smart Tools and Applications in Computer Graphics. Eurographics Association, 2019.
Copyright (c) 2019 DFKI GmbH.
MIT License, see the LICENSE file.