AVI 2022 Demo

This version of the APIs is the same version that is used in the frontend presented at the AVI 2022 conference in Fascati, Rome (Italy) (see the appropriate release page). As soon as the paper will be published, it will be available in the ACM Digital Library at the following DOI link: 10.1145/3531073.3534464.

Please note that, differently from the frontend, this is not considered as a pre-release: the structure and methods of the API are already pretty stable, and future work will probably be centered on the actual AI model and/or on documentation.

Furthermore, please note that to make this API work, the set of AI models to be used for predictions must be put in the main folder under the models/ folder. An initial version of such models, used in the AVI 2022 demo, is provided here in the models.zip file.