/DAPS

Official Repository of Dense 2D-3D Indoor Prediction with Sound via Aligned Cross-Modal Distillation (ICCV 2023)

Primary LanguagePythonMIT LicenseMIT

DAPS

Official repository of Dense 2D-3D Indoor Prediction with Sound via Aligned Cross-Modal Distillation (ICCV 2023)

example

[Project Page]

DAPS (Dense Auditory Prediction of Surroundings) benchmark

dataset

We curated the DAPS benchmark based on Matterport3D and SoundSpaces. Tools and detailed instructions for preparing the data is available here.

SAM (Spatial Alignment via Matching) distillation framework

arch

The code for our model and evaluation will be released shortly.

Citation

If you find our work useful in your research, please consider citing:

@InProceedings{
    author    = {Yun, Heeseung and Na, Joonil and Kim, Gunhee},
    title     = {Dense 2D-3D Indoor Prediction with Sound via Aligned Cross-Modal Distillation},
    booktitle = {ICCV},
    year      = {2023}
}