Add pipeline tag and model card content

README.md

@@ -1,9 +1,22 @@
 ---
-license: apache-2.0
+base_model:
+- TencentARC/Pixal3D
 datasets:
 - AlayaLab/Worldsculpt_data
 language:
 - en
-base_model:
-- TencentARC/Pixal3D
----
+license: apache-2.0
+pipeline_tag: image-to-3d
+---
+
+# WorldSculpt
+
+This repository contains the model presented in [WorldSculpt: Generating Compositional Worlds from Grounded Videos](https://huggingface.co/papers/2609.05416).
+
+WorldSculpt generates compositional 3D mesh scenes from posed multi-view images with instance masks and 3D boxes, enabling complex scenes with up to hundreds of objects.
+
+- **Project page:** [https://alaya-lab.github.io/WorldSculpt/](https://alaya-lab.github.io/WorldSculpt/)
+- **Code:** [https://github.com/AlayaLab/WorldSculpt](https://github.com/AlayaLab/WorldSculpt)
+- **Dataset:** [AlayaLab/Worldsculpt_data](https://huggingface.co/datasets/AlayaLab/Worldsculpt_data)
+
+For setup and usage instructions, please see the [GitHub repository](https://github.com/AlayaLab/WorldSculpt).