AVLEN: Audio-Visual-Language Embodied Navigation in 3D Environments - Supplementary Data

Introduction

In this zip, we release the auxiliary data that is beneficial to execute the implementation of AVLEN described in our paper AVLEN: Audio-Visual-Language Embodied Navigation in 3D Environments by Sudipta Paul, Amit K Roy-Chowdhury, and Anoop Cherian, NeurIPS, 2022.

At a Glance

• The size of the unzipped data is 4.6G
• The unzipped folder contains: (i) a README.md file and (ii) ./AVLEN-data folder. The latter contains the following zip files. Please see the AVLEN code to see how to unzip these files into their respective folders.
  • ckpt.119.pth  -- 61M  
  • connectivity.zip -- 1.4M 
  • pretrained_weights.zip -- 1.7G
  • ResNet-152-imagenet.zip -- 2.9G
  • semantic_audionav_dialog_approx.zip -- 2.7M
  • soundspaces.zip -- 479K
  • speaker_model_weights.zip -- 51M

Other Resources

For the implementation of AVLEN that uses the data shared here, please visit MERL TR2022-131.

Citation

If you use AVLEN in your research, please cite our paper:

@InProceedings{paul2022avlen, 
    title={AVLEN: Audio-Visual-Language Embodied Navigation in 3D Environments}, 
    booktitle={Advances in Neural Information Processing Systems},
    author={Paul, Sudipta and Roy-Chowdhury, Amit and Cherian, Anoop}, 
    volume={35}, 
    pages={6236--6249}, 
    year={2022} 
}

Copyright and License

The AVLEN dataset is released under CC-BY-SA-4.0 license.

All data:

Created by Mitsubishi Electric Research Laboratories (MERL), 2023

SPDX-License-Identifier: CC-BY-SA-4.0