Improve model card: Add paper link, abstract, usage, and citation; refine metadata

This PR significantly enhances the model card for `SS4D: Native 4D Generative Model via Structured Spacetime Latents` by:

- **Updating the title** to include the full paper name for better clarity.
- **Removing `library_name: ss4d`** from the metadata, as it refers to a custom implementation and does not correspond to a standard Hugging Face library intended for automated code snippets on the Hub, as per guidelines.
- **Adding a direct link to the paper** on the Hugging Face Hub (`https://huggingface.co/papers/2512.14284`).
- **Including the paper's abstract** for a concise overview of the model.
- **Embedding the teaser image** from the GitHub repository for visual context.
- **Adding detailed installation instructions** from the GitHub README.
- **Including a sample usage snippet** (`python test.py`) as found in the GitHub README.
- **Adding the academic citation** for the work.
- **Maintaining the existing project page and code links**, along with `language: en`, `license: mit`, and `pipeline_tag: image-to-3d`.

These changes aim to make the model card more comprehensive and user-friendly, providing essential information directly on the Hub.

README.md

@@ -1,14 +1,59 @@
 ---
-library_name: ss4d
-pipeline_tag: image-to-3d
-license: mit
 language:
 - en
+license: mit
+pipeline_tag: image-to-3d
 ---
-# SS4D
 
-Official model weights for *SS4D: Native 4D Generative Model via Structured Spacetime Latents*
+# SS4D: Native 4D Generative Model via Structured Spacetime Latents
+
+Official model weights for the paper [SS4D: Native 4D Generative Model via Structured Spacetime Latents](https://huggingface.co/papers/2512.14284).
+
+[Paper](https://huggingface.co/papers/2512.14284) | [Project Page](https://lizb6626.github.io/SS4D/) | [Code](https://github.com/Lizb6626/SS4D/)
+
+<p align="center"><img src="https://github.com/Lizb6626/SS4D/raw/main/assets/teaser.jpg" width="100%"></p>
+
+We present SS4D, a native 4D generative model that synthesizes dynamic 3D objects directly from monocular video. Unlike prior approaches that construct 4D representations by optimizing over 3D or video generative models, we train a generator directly on 4D data, achieving high fidelity, temporal coherence, and structural consistency. At the core of our method is a compressed set of structured spacetime latents. Specifically, (1) To address the scarcity of 4D training data, we build on a pre-trained single-image-to-3D model, preserving strong spatial consistency. (2) Temporal consistency is enforced by introducing dedicated temporal layers that reason across frames. (3) To support efficient training and inference over long video sequences, we compress the latent sequence along the temporal axis using factorized 4D convolutions and temporal downsampling blocks. In addition, we employ a carefully designed training strategy to enhance robustness against occlusion and motion blur, leading to high-quality generation. Extensive experiments show that SS4D produces spatio-temporally consistent 4D objects with superior quality and efficiency, significantly outperforming state-of-the-art methods on both synthetic and real-world datasets.
+
+## Installation
+
+To set up the environment, follow the instructions from the GitHub repository:
+
+```sh
+git clone --recurse-submodules git@github.com:Lizb6626/SS4D.git
+cd SS4D
+./setup.sh --new-env --basic --xformers --flash-attn --diffoctreerast --spconv --mipgaussian --kaolin --nvdiffrast
+```
+
+Alternatively, you can follow the detailed installation guidance provided by [TRELLIS](https://github.com/microsoft/TRELLIS/tree/main?tab=readme-ov-file#-installation).
+
+## Usage
+
+We provide a minimal example to run our method on the provided examples:
+
+```
+python test.py
+```
 
-Project page: https://lizb6626.github.io/SS4D/
+## Citation
 
-Code: https://github.com/Lizb6626/SS4D/
+```bibtex
+@article{li2025ss4d,
+  author = {Li, Zhibing and Zhang, Mengchen and Wu, Tong and Tan, Jing and Wang, Jiaqi and Lin, Dahua},
+  title = {SS4D: Native 4D Generative Model via Structured Spacetime Latents},
+  year = {2025},
+  issue_date = {December 2025},
+  publisher = {Association for Computing Machinery},
+  address = {New York, NY, USA},
+  volume = {44},
+  number = {6},
+  issn = {0730-0301},
+  url = {https://doi.org/10.1145/3763302},
+  doi = {10.1145/3763302},
+  journal = {ACM Trans. Graph.},
+  month = dec,
+  articleno = {244},
+  numpages = {12},
+  keywords = {4D generation, 3D generation, animation, generative model}
+}
+```