Update dataset card with metadata and project links (#2)

- Update dataset card with metadata and project links (af2c79422f423c10b7f29d6ee1f52c7abb3b40bc)


Co-authored-by: Niels Rogge <nielsr@users.noreply.huggingface.co>

README.md

@@ -1,4 +1,7 @@
 ---
+license: cc-by-sa-4.0
+task_categories:
+- image-to-3d
 dataset_info:
   features:
   - name: uid
@@ -123,8 +126,33 @@ configs:
     path: data/val-*
   - split: test
     path: data/test-*
-license: cc-by-sa-4.0
 ---
-# Dataset Card
 
-This is the training dataset for https://arxiv.org/abs/2603.05888.
+# PixARMesh Training Dataset
+
+[**Project Page**](https://mlpc-ucsd.github.io/PixARMesh/) | [**Paper**](https://arxiv.org/abs/2603.05888) | [**GitHub**](https://github.com/mlpc-ucsd/PixARMesh)
+
+This repository contains the training dataset for **PixARMesh**, a method to autoregressively reconstruct complete 3D indoor scene meshes directly from a single RGB image. Unlike prior methods that rely on implicit signed distance fields, PixARMesh jointly predicts object layout and geometry within a unified model, producing coherent and artist-ready meshes in a single forward pass.
+
+## Dataset Preparation
+
+According to the official repository, you should flatten the dataset to ensure uniform instance sampling across scenes:
+
+```bash
+python -m scripts.flatten_dataset
+```
+
+This prevents instances from scenes with many objects from being under-sampled during training.
+
+## Citation
+
+If you find PixARMesh useful in your research, please consider citing:
+
+```bibtex
+@article{zhang2026pixarmesh,
+  title={PixARMesh: Autoregressive Mesh-Native Single-View Scene Reconstruction},
+  author={Zhang, Xiang and Yoo, Sohyun and Wu, Hongrui and Li, Chuan and Xie, Jianwen and Tu, Zhuowen},
+  journal={arXiv preprint arXiv:2603.05888},
+  year={2026}
+}
+```