/NoPrior_MultiSSL

Official Repository for "Learning to Visually Localize Sound Sources from Mixtures without Prior Source Knowledge" (CVPR 2024)

Primary LanguagePythonMIT LicenseMIT

Learning to Visually Localize Sound Sources from Mixtures without Prior Source Knowledge

Official Repository for "Learning to Visually Localize Sound Sources from Mixtures without Prior Source Knowledge".

Accepted at CVPR 2024:

[Paper] [arXiv]

Citation

If you find this work useful for your research, please cite the paper:

@inproceedings{kim2024learning,
    author    = {Kim, Dongjin and Um, Sung Jin and Lee, Sangmin and Kim, Jung Uk},
    title     = {Learning to Visually Localize Sound Sources from Mixtures without Prior Source Knowledge},
    booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
    year      = {2024},
}