--- title: DiffusionGS Scene Generation emoji: ๐Ÿ  colorFrom: blue colorTo: indigo sdk: static app_file: index.html pinned: false short_description: Image-to-3D scenes, videos and point-cloud downloads models: - CaiYuanhao/DiffusionGS datasets: - CaiYuanhao/DiffusionGS tags: - image-to-3d - image-to-video - gaussian-splatting - diffusiongs --- # DiffusionGS ยท Scene Generation Generate a 3D scene from one reference image. Upload an image or choose an example, click **Generate scene & video**, then preview the generated views and video and download the Gaussian point cloud. [Object demo](https://huggingface.co/spaces/CaiYuanhao/DiffusionGS-Object) ยท [GitHub](https://github.com/caiyuanhao1998/Open-DiffusionGS) ยท [Models](https://huggingface.co/CaiYuanhao/DiffusionGS) ยท [Paper](https://arxiv.org/abs/2411.14384) ![Scene generation interface](img/demos/scene_demo.png) This Space embeds the Scene Gradio interface hosted at the shared [ZeroGPU backend](https://caiyuanhao-diffusiongs-object.hf.space/scene). It runs the same `scene_demo.py` as the local demo, including its examples, controls, video preview and downloads.