/RedCore

Primary LanguagePython

RedCore

The paper "RedCore: Relative Advantage Aware Cross-modal Representation Learning for Missing Modalities with Imbalanced Missing Rates" is accepted by AAAI 2024.

The features extracted from IEMOCAP, CMU-MOSEI, and MSP-IMPROV can be downloaded from: https://drive.google.com/drive/folders/1OPiw4XFTzTnoKxT14jtSif5XVN8Qe7l4?usp=sharing

Please follow the steps below to run the code.

  1. Down load the features from: https://drive.google.com/drive/folders/1OPiw4XFTzTnoKxT14jtSif5XVN8Qe7l4?usp=sharing
  2. Set the path to the features in the json files under folder: data/config/
  3. Run the code by: bash run.sh

Acknowledgement: the implementation is based on MMIN: https://github.com/AIM3-RUC/MMIN/tree/master.

If you find our work is helpful in your research, please cite our paper:

@article{sun2023redcore, title={RedCore: Relative Advantage Aware Cross-modal Representation Learning for Missing Modalities with Imbalanced Missing Rates}, author={Sun, Jun and Zhang, Xinxin and Han, Shoukang and Ruan, Yu-ping and Li, Taihao}, journal={arXiv preprint arXiv:2312.10386}, year={2023} }