MoGeInference - ComfyUI Built-in Node Documentation

Run MoGe on a single image to estimate depth and geometry.

Run MoGe on a single image to estimate depth and geometry. This node processes an input image through the MoGe model to generate a 3D point cloud, depth map, camera intrinsics, a mask, and surface normals.

Inputs

Parameter	Description	Data Type	Required	Range
moge_model	The MoGe model to use for inference.	MOGE_MODEL	Yes	N/A
image	The input image for depth and geometry estimation. Only the first three color channels (RGB) are used.	IMAGE	Yes	N/A
resolution_level	Controls the processing resolution. 0 is fastest, 9 provides the most detail. (default: 9)	INT	Yes	0 to 9
fov_x_degrees	(Advanced) Horizontal field of view of the source camera in degrees. Sets the focal length used to unproject the depth map into 3D. Set to 0.0 to automatically recover the field of view from the predicted points. (default: 0.0)	FLOAT	Yes	0.0 to 170.0
batch_size	Images per inference call. Lower this value if you run out of memory on a long video or image set. (default: 4)	INT	Yes	1 to 64
force_projection	(Advanced) Forces projection of the predicted points. (default: True)	BOOLEAN	Yes	True/False
apply_mask	(Advanced) Sets masked-out (sky or invalid) pixels to infinity in the points and depth outputs so meshing tools can ignore them. Disable to keep the raw predicted geometry everywhere; the mask is still returned separately. (default: True)	BOOLEAN	Yes	True/False

Note: When the input image contains more frames than batch_size, the node processes them in multiple inference calls and combines the results into a single output geometry.

Outputs

Output Name	Description	Data Type
moge_geometry	A dictionary containing the estimated geometry. It includes the original image, and may contain points (3D point cloud), depth (depth map), intrinsics (camera intrinsics matrix), mask (mask identifying valid pixels), and normal (surface normals).	MOGE_GEOMETRY

This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! Edit on GitHub

---

Source fingerprint (SHA-256): 59f6b8b1ab65147a47f5dc7ebee7b965a5ab37c6a0843a2c80d50c767ad98db4