Multi-Class Textual-Inversion Secretly Yields a Semantic-Agnostic Classifier (WACV 2025)

NOTICE

The code repo will be updated soon. For now it's just for arxiv connection.

Training Guidance

Due to person data privacy, we didn't include photos.

You should collect Lecun's photo and put under

./data/person/images_class_n_shots/4/

Then you can start training by

./bash_files/person_train.sh