|
--- |
|
license: cc-by-nc-4.0 |
|
viewer: false |
|
extra_gated_prompt: >- |
|
Explicit consent is given from Intelligent Interaction Group for academic research purposes only. |
|
The rights to the annotation of the MER dataset belong to Intelligent Interaction Group. |
|
No legal claims of any kind can be derived from accepting or using the dataset. |
|
Intelligent Interaction Group shall not be held liable for any damage resulting from accessing or using the dataset or any other files provided. |
|
The licensee is not permitted to distribute the dataset or any derivative files (such as annotation or labeling files) to third parties, |
|
nor may they modify the dataset without prior written consent from Intelligent Interaction Group. |
|
|
|
extra_gated_fields: |
|
Team Name: text |
|
Team Members (separated by commas): text |
|
Organization: text |
|
Country: country |
|
Specific Date: date_picker |
|
I want to use this dataset for: |
|
type: select |
|
options: |
|
- Research |
|
- Education |
|
- label: Other |
|
value: other |
|
I agree to use this dataset for non-commercial use ONLY: checkbox |
|
language: |
|
- en |
|
--- |
|
|
|
 |
|
|
|
# MER2025 Dataset |
|
|
|
Welcome to the official dataset for the **[MER2025 Challenge@ACM MM](https://zeroqiaoba.github.io/MER2025-website/)**! |
|
|
|
MER2025 marks the third edition of our **Multimodal Emotion Recognition (MER)** series of challenges, aiming to bring together the affective computing community to explore emerging trends and future directions. This year, MER2025 focuses on the theme: |
|
### **"Affective Computing Meets Large Language Models"** |
|
|
|
We seek to shift from traditional categorical emotion recognition frameworks, which rely on predefined taxonomies, towards **LLM-driven generative methods**, offering innovative solutions for more accurate and interpretable emotion understanding. |
|
|
|
Baseline code is available here: [MERTools for MER2025](https://github.com/zeroQiaoba/MERTools/tree/master/MER2025) |
|
|
|
--- |
|
|
|
## 🌐 News |
|
|
|
- **April 30, 2025**: MER25 Challenge and MRAC25 Workshop website launched! |
|
|
|
--- |
|
|
|
## 🔥 Challenge Tracks |
|
|
|
MER2025 features four tracks: |
|
|
|
### Track 1. MER-SEMI |
|
Enhancing categorical emotion recognition through **semi-supervised learning**. |
|
Participants are encouraged to leverage methods such as masked auto-encoders and contrastive learning to improve performance using both labeled and unlabeled data. |
|
|
|
- Predict one label from six categories: **worried, happy, neutral, angry, surprise, sad**. |
|
|
|
### Track 2. MER-FG |
|
Moving beyond basic emotions to **fine-grained emotional states**[1, 2]. |
|
Participants can use LLMs to predict **any emotion labels**, capturing more nuanced emotional expressions. |
|
|
|
### Track 3. MER-DES |
|
Integrating **multimodal clues** for **explainable emotion recognition**[2, 3]. |
|
Instead of relying solely on emotion words, participants must submit **both multimodal evidence and emotion labels**, enhancing interpretability. |
|
|
|
### Track 4. MER-PR |
|
Exploring the relationship between **emotion recognition and personality prediction**. |
|
Participants are encouraged to use techniques like multi-task learning to determine whether better emotion recognition improves personality prediction. |
|
|
|
- Related references: |
|
- [1] OV-MER: Towards Open-Vocabulary Multimodal Emotion Recognition. [arXiv:2410.01495](https://arxiv.org/abs/2410.01495) |
|
- [2] AffectGPT: A New Dataset, Model, and Benchmark for Emotion Understanding with Multimodal Large Language Models. [arXiv:2501.16566](https://arxiv.org/abs/2501.16566) |
|
- [3] Explainable Multimodal Emotion Recognition. [arXiv:2306.15401v5](https://arxiv.org/abs/2306.15401) |
|
|
|
|
|
--- |
|
|
|
## 📂 Dataset Access |
|
|
|
- **Academic use only**: Access to the dataset requires signing an EULA (End-User License Agreement). |
|
- **No redistribution**: Uploading the dataset to any public platform or modifying it is prohibited. |
|
- **Approval required**: After submitting the EULA on Hugging Face, access will be granted upon approval. |
|
|
|
## 📥 Access Instructions |
|
|
|
After your access request has been approved, the dataset download instructions will be provided in the file: |
|
**`README_AFTER_APPROVAL.md`** |
|
|
|
Please carefully review the LICENSE and EULA requirements before submitting your access request. |
|
|
|
--- |
|
|
|
## 🧪 Result Submission |
|
|
|
- For **MER-SEMI**, **MER-FG**, and **MER-DES**: |
|
- Submit predictions for **20,000** selected candidate samples (subsets of the original 124k unlabeled samples). |
|
- For **MER-PR**: |
|
- Submit predictions directly on the provided test set. |
|
|
|
Each track has different prediction formats and targets as described above. |
|
|
|
--- |
|
|
|
## 📝 Paper Submission |
|
|
|
All participants are highly encouraged to submit a paper describing their proposed solutions to the **Workshop Proceedings** or **Challenge Proceedings**, held as part of ACM Multimedia 2025. |
|
|
|
|
|
- **Baseline paper**: https://arxiv.org/abs/2504.19423 |
|
- **Baseline code**: https://github.com/zeroQiaoba/MERTools/tree/master/MER2025 |
|
- **Contact email**: [email protected]; [email protected] |
|
--- |
|
|
|
## 📅 Important Dates |
|
|
|
| Event | Date | |
|
| :--- | :--- | |
|
| Data, baseline paper & code available | April 30, 2025 | |
|
| Results submission start | June 26, 2025 | |
|
| Results submission deadline |July 13, 2025, noon UTC |
|
| Paper submission deadline | July 20, 2025 | |
|
| Paper acceptance notification | August 1, 2025 | |
|
| Camera-ready paper deadline | August 11, 2025 | |
|
| MRAC25 Workshop@ACM MM (Dublin, Ireland) | Oct 27-31, 2025 | |
|
|
|
<!-- > ⏰ All deadlines are **23:59 Anywhere on Earth (AoE)**. --> |
|
|
|
--- |
|
|
|
## 📫 Contact |
|
|
|
For any questions, collaborations, or issues: |
|
> **[email protected]** |
|
|
|
You can also raise an issue in the [MERTools GitHub repository](https://github.com/zeroQiaoba/MERTools/tree/master/MER2025). |
|
|
|
--- |
|
|
|
# 🛡️ License |
|
|
|
© 2025 Intelligent Interaction Group. All rights reserved. |
|
|
|
This dataset is made available for **academic research purposes only** under a custom End-User License Agreement (EULA) and a [customized CC BY-NC 4.0 License](https://github.com/zeroQiaoba/MERTools/blob/master/MER2025/LICENSE) with additional usage restrictions. |
|
|
|
**Users must:** |
|
- Sign and submit the EULA. |
|
- Obtain approval before accessing the data. |
|
- Use the dataset solely for non-commercial academic research purposes. |
|
- Not modify, redistribute, or publish the dataset or any derivative works without prior written consent. |
|
|
|
📄 Please refer to the [LICENSE file](https://github.com/zeroQiaoba/MERTools/blob/master/MER2025/LICENSE) for full license and usage details. |
|
|
|
--- |