bibekram/emotion_multimodal

Multimodal Facial Emotion Recognition Dataset (EEG + Thermal + Digital Image)

📌 Summary

This dataset contains multimodal recordings of 50 subjects across six emotional states:

• Happy
• Sad
• Anger
• Fear
• Surprise
• Neutral (Normal)

Each emotion is captured in three different modalities:

1. EEG Data – 4-channel EEG signals recorded during emotion elicitation.
2. Thermal Images – infrared thermal facial images.
3. Digital Images – standard RGB facial images.

This dataset was created for the research project:
“Multimodal Facial Emotion Recognition by Correlating EEG, Thermal, and Digital Image Data.”

---

📂 Dataset Structure

EEG/ ├── happy/ ├── sad/ ├── anger/ ├── fear/ ├── surprise/ └── neutral/ Thermal/ ├── happy/ ├── sad/ ├── anger/ ├── fear/ ├── surprise/ └── neutral/ Digital_Image/ ├── happy/ ├── sad/ ├── anger/ ├── fear/ ├── surprise/ └── neutral/

• Each subfolder contains subject-wise data samples for the corresponding emotion.
• Total size: 50 subjects × 6 emotions × 3 modalities.
• EEG files are provided in text/CSV format (per subject per emotion).
• Thermal and Digital images are stored in standard image formats (JPG/PNG).

---

🔧 Usage

You can load this dataset with Hugging Face datasets:

from datasets import load_dataset

dataset = load_dataset("bibekram/emotion_multimodal")
print(dataset)

💡 Applications

This dataset can be used for:

Multimodal Emotion Recognition

Human-Computer Interaction (HCI)

Affective Computing

Brain–Computer Interface (BCI)

Cross-modal feature correlation research

📑 License

This dataset is released under the CC BY 4.0 License.
You are free to use, modify, and share it, provided proper attribution is given.

🙌 Citation

If you use this dataset, please cite:

Bibek Ram (2025). Multimodal Emotion Recognition Dataset (EEG + Thermal + Digital Image). Hugging Face.

BibTeX:

@dataset{bibekram2025multimodal,
  author = {Bibek Ram},
  title = {Multimodal Emotion Recognition Dataset (EEG + Thermal + Digital Image)},
  year = {2025},
  publisher = {Hugging Face},
  url = {https://huggingface.co/datasets/bibekram/emotion_multimodal}
}