paper

The code for our paper "MIT-FRNet: Modality-Invariant Temporal Representation Learning-based Feature Reconstruction Network for Missing Modalities".

tool

Python 3.6

datasets

1. MOSI:

paper title: Transformer-based Feature Reconstruction Network for Robust Multimodal Sentiment Analysis

data load: https://drive.google.com/file/d/1-7UOqVxdpTA_9auZLtZ93oUmhrajaZlT/view?usp=drive_link

2. SIMS：

paper title: CH-SIMS: A Chinese Multimodal Sentiment Analysis Dataset with Fine-grained Annotations of Modality

data load: https://drive.google.com/file/d/10WrmbrzIVvaeEFsB0GMW4Hn1QSg-Ll8j/view?usp=drive_link

baselines

1. TFN https://github.com/A2Zadeh/TensorFusionNetwork

2. MulT https://github.com/yaohungt/Multimodal-Transformer

3. MISA https://github.com/declare-lab/MISA

4. TFR-Net https://github.com/Columbine21/TFR-Net

cite

please cite our paper if you find our work useful for your research:

{

---

}

note

If you have any question, feel free to contact us.