Add model card and pipeline tag

README.md

@@ -1,3 +1,34 @@
----
-license: apache-2.0
----
+---
+license: apache-2.0
+pipeline_tag: image-to-3d
+---
+
+# Uni3R: Unified 3D Reconstruction and Semantic Understanding via Generalizable Gaussian Splatting from Unposed Multi-View Images
+
+Uni3R is a novel feed-forward framework that jointly reconstructs a unified 3D scene representation enriched with open-vocabulary semantics, directly from unposed multi-view images. It leverages a Cross-View Transformer to integrate information across arbitrary multi-view inputs and regresses a set of 3D Gaussian primitives endowed with semantic feature fields.
+
+[**Project Page**](https://horizonrobotics.github.io/robot_lab/uni3R/) | [**Paper (arXiv:2508.03643)**](https://arxiv.org/abs/2508.03643) | [**GitHub Code**](https://github.com/HorizonRobotics/Uni3R)
+
+## Key Features
+- **Feed-forward Reconstruction**: Jointly handles 3D reconstruction and semantic interpretation without requiring costly per-scene optimization.
+- **Unified Representation**: Facilitates high-fidelity novel view synthesis, open-vocabulary 3D semantic segmentation, and depth prediction in a single pass.
+- **Unposed Inputs**: Robustly integrates information across arbitrary multi-view inputs without pre-defined camera poses.
+- **Generalizable**: Establishes state-of-the-art performance on benchmarks like RE10K and ScanNet.
+
+## Usage
+For detailed installation and usage instructions, please refer to the official [GitHub repository](https://github.com/HorizonRobotics/Uni3R). The repository provides scripts for training and evaluation on 2, 8, and 16 views.
+
+## Citation
+If you find this work useful in your research, please consider citing:
+
+```bibtex
+@misc{sun2025uni3runified3dreconstruction,
+      title={Uni3R: Unified 3D Reconstruction and Semantic Understanding via Generalizable Gaussian Splatting from Unposed Multi-View Images}, 
+      author={Xiangyu Sun and Haoyi Jiang and Liu Liu and Seungtae Nam and Gyeongjin Kang and Xinjie Wang and Wei Sui and Zhizhong Su and Wenyu Liu and Xinggang Wang and Eunbyung Park},
+      year={2025},
+      eprint={2508.03643},
+      archivePrefix={arXiv},
+      primaryClass={cs.CV},
+      url={https://arxiv.org/abs/2508.03643}, 
+}
+```