Update dataset card for MinD-3D++ paper and add task category (#2)

- Update dataset card for MinD-3D++ paper and add task category (989739bc8bc2526b3dc36b2aab59c086b356d4c1)


Co-authored-by: Niels Rogge <[email protected]>

README.md

@@ -1,14 +1,25 @@
 ---
 license: apache-2.0
+task_categories:
+- image-to-3d
+tags:
+- fmri
+- 3d-reconstruction
+- neuroscience
+- brain-decoding
 ---
 
-# [ECCV 2024] MinD-3D: Reconstruct High-quality 3D objects in Human Brain
+# fMRI-Shape Dataset: A Component of the fMRI-3D Dataset for MinD-3D++
 
-[![ArXiv](https://img.shields.io/badge/ArXiv-2312.07485-b31b1b.svg?logo=arXiv)](https://arxiv.org/abs/2312.07485)
-[![Github](https://img.shields.io/badge/Github-MinD_3D-blue.svg?logo=Github)](https://github.com/JianxGao/MinD-3D)
+This repository contains the `fMRI-Shape` dataset, a component of the comprehensive fMRI-3D dataset introduced and utilized in the paper [MinD-3D++: Advancing fMRI-Based 3D Reconstruction with High-Quality Textured Mesh Generation and a Comprehensive Dataset](https://huggingface.co/papers/2409.11315). This work builds upon the initial "MinD-3D" research.
+
+The fMRI-3D dataset consists of two components: `fMRI-Shape` (this dataset) and [fMRI-Objaverse](https://huggingface.co/datasets/Fudan-fMRI/fMRI-Objaverse). Both datasets are designed to advance 3D visual reconstruction from fMRI data.
+
+**Project Page:** https://jianxgao.github.io/MinD-3D
+**Code (GitHub):** https://github.com/JianxGao/MinD-3D
 
 ## Overview
-MinD-3D aims to reconstruct high-quality 3D objects based on fMRI data. 
+MinD-3D (and MinD-3D++) aims to reconstruct high-quality 3D objects based on fMRI data. This `fMRI-Shape` dataset contains fMRI recordings and corresponding 3D object stimuli used in this research.
 
 ## Repository Structure
 - **annotations**: Contains metadata and annotations related to the fMRI data for each subject.
@@ -20,12 +31,11 @@ MinD-3D aims to reconstruct high-quality 3D objects based on fMRI data.
 - **raw_data**: Raw fMRI data collected directly from the imaging machine.
 - **npy_data**: Processed data. We utilized fMRIPrep and the methodologies described in our paper to derive and store the data in NumPy format (.npy).
 
-
 ## Citation
 
-If you find our paper useful for your research and applications, please cite using this BibTeX:
+If you find our work and datasets useful for your research and applications, please cite the respective papers:
 
-```
+```bibtex
 @misc{gao2023mind3d,
   title={MinD-3D: Reconstruct High-quality 3D objects in Human Brain}, 
   author={Jianxiong Gao and Yuqian Fu and Yun Wang and Xuelin Qian and Jianfeng Feng and Yanwei Fu},
@@ -34,4 +44,16 @@ If you find our paper useful for your research and applications, please cite usi
   archivePrefix={arXiv},
   primaryClass={cs.CV}
 }
+```
+
+```bibtex
+@misc{gao2025mind3dadvancingfmribased3d,
+      title={MinD-3D++: Advancing fMRI-Based 3D Reconstruction with High-Quality Textured Mesh Generation and a Comprehensive Dataset}, 
+      author={Jianxiong Gao and Yanwei Fu and Yuqian Fu and Yun Wang and Xuelin Qian and Jianfeng Feng},
+      year={2025},
+      eprint={2409.11315},
+      archivePrefix={arXiv},
+      primaryClass={cs.CV},
+      url={https://arxiv.org/abs/2409.11315}, 
+}
 ```