/EZ-AVGZL

Official Codebase of "Audio-visual Generalized Zero-shot Learning the Easy Way" (ECCV 2024)

Audio-visual Generalized Zero-shot Learning the Easy Way

We will release our code soon!

EZ-AVGZL is a novel framework that can align audio-visual embeddings with transformed text representations for Easy Audio-Visual Generalized Zero-shot Learning,

Audio-visual Generalized Zero-shot Learning the Easy Way
Shentong Mo, Pedro Morgado
ECCV 2024.

EZ-AVGZL Illustration

Citation

If you find this repository useful, please cite our paper:

@inproceedings{mo2024ezavgzl,
  title={Audio-visual Generalized Zero-shot Learning the Easy Way},
  author={Mo, Shentong and Morgado, Pedro},
  booktitle={Proceedings of European Conference on Computer Vision},
  year={2024}
}