Spaces:
Running
Running
| title: LingBot 3D | |
| short_description: Streaming 3D scene reconstruction from short videos. | |
| colorFrom: blue | |
| colorTo: green | |
| sdk: gradio | |
| app_file: app.py | |
| python_version: 3.10.13 | |
| pinned: false | |
| license: apache-2.0 | |
| startup_duration_timeout: 1h | |
| models: | |
| - robbyant/lingbot-map | |
| preload_from_hub: | |
| - robbyant/lingbot-map lingbot-map.pt | |
| # LingBot 3D | |
| Turn a short video into a 3D scene with LingBot-Map. | |
| - uses the upstream `lingbot_map` package directly | |
| - downloads checkpoints from `robbyant/lingbot-map` | |
| - runs the SDPA fallback path instead of FlashInfer | |
| - caps inputs to short clips so the app fits a shared ZeroGPU workflow | |
| - exports a browser-friendly `.glb` scene plus a zipped artifact bundle | |
| ## Current Limits | |
| - short demos only | |
| - default frame cap: 24 frames | |
| - preview uses a static embedded `viser` scene, not the live local `viser` server | |
| - the app uses the upstream default checkpoint `lingbot-map.pt` | |
| ## Local Sanity Check | |
| If you want to import the app locally without downloading the checkpoint at startup: | |
| ```bash | |
| LINGBOT_SPACE_SKIP_MODEL_LOAD=1 python app.py | |
| ``` | |
| ## Upstream | |
| - GitHub: https://github.com/Robbyant/lingbot-map | |
| - Model: https://huggingface.co/robbyant/lingbot-map | |