Domain adaptation for action recognition

Warning

This tutorial is a stub. The example in this folder is complete and runnable – see the README and run the command below – but the narrated walkthrough has not been written yet. Tracking issue: the tutorial is marked draft: true in metadata.yml.

Transfer an action recogniser between egocentric video domains. Same adversarial adaptation pipeline as the digits tutorial, with a 3D-CNN feature extractor and video data loaders.

To run the full example now:

python main.py --cfg configs/EPIC-D12D2-DANN.yaml --devices 1

Setup

Placeholder so the gallery builds. Replace this file using source/_shared/templates/plot.py.template.

print("See main.py in this folder for the runnable example.")