ComfyUI Node
Plot Camera
A ComfyUI node in SplatKit with 9 inputs and 5 outputs.
Plot Camera
- panorama
- moge_model
- control_video
- control_mask
- condition_dir
- camera_preview
- rail_json
◄anchors0, 0, 0
0.6, 0.1, 1.5
-0.4, 0.2, 3.0
0.3, 0.0, 4.5►
◄orientationlook_forward►
◄length81►
◄moge_level6►
◄dataset_dir►
◄moge_ckpt▾►
◄look_at_target0, 0, 3►
CategorySplatKit
Inputs (9)
| Name | Type | Default | Description |
|---|---|---|---|
| panorama | IMAGE | — | |
| anchors | STRING | 0, 0, 0 0.6, 0.1, 1.5 -0.4, 0.2, 3.0 0.3, 0.0, 4.5 | Fly-through points, in the SAME units as the geometry overlay. Frame: +Z forward/into the pano, +X right, +Y up; origin = start camera. The camera goes EXACTLY to each point you place against the cloud (WYSIWYG). One 'x,y,z' per line, or JSON [[x,y,z],...]. Need at least 2 points. |
| orientation | COMBO | look_forward | look_forward = camera faces the path tangent (default). look_at_target = every frame aims at the SAME shared point -- set it in the look_at_target widget or drag the orange 'look' marker in the editor. per_point_look = each anchor has its own draggable look target; the aim sweeps between them (set the target interactively in the editor). |
| length | INT | 819–257 | Number of frames. MUST match the Wan Conditioning length (81). |
| moge_level | INT | 60–9 | Geometry detail: the MoGe depth inference resolution level (0-9). Higher = sharper depth edges / thin structures but slower (~2x cost from 0 to 9); lower = faster, softer geometry. 6 is a balanced default -- drop to 3-4 for quick anchor placement, raise to 9 for a final sharp mesh. Output resolution is unaffected; this only sets how much detail the depth network sees per panorama sub-view. |
| dataset_diropt | STRING | Wire the Dataset Project node here. When set, condition/ is written under it; otherwise it falls back to a default output folder. | |
| moge_ckptopt | COMBO | MoGe checkpoint from ComfyUI/models/MoGe. 'auto (download)' fetches 'Ruicheng/moge-vitl' into that folder on first use (~1.2GB). Drop your own model.pt in models/MoGe to pick it here, or wire a MoGe Model Loader node into 'moge_model'. | |
| moge_modelopt | MOGE_MODEL | Optional: a pre-loaded MoGe model from the MoGe Model Loader node. Overrides moge_ckpt; load once and reuse across nodes. | |
| look_at_targetopt | STRING | 0, 0, 3 | The single world point ALL frames aim at when orientation = look_at_target. 'x, y, z' in the SAME literal units as the anchors / geometry overlay. Draggable in the editor (the orange 'look' marker). Ignored by the other orientation modes. |
Outputs (5)
| Name | Type | Description |
|---|---|---|
| control_video | IMAGE | — |
| control_mask | IMAGE | — |
| condition_dir | STRING | — |
| camera_preview | IMAGE | — |
| rail_json | STRING | — |