Pixal3DConditioning - ComfyUI Built-in Node Documentation

The Pixal3DConditioning node prepares image conditioning for the Trellis2 3D generation pipeline. It uses a DINOv3 vision model to extract visual features from the input image at two resolutions (512 and 1024), then organizes them into per-stage feature maps that can be optionally enhanced by a NAF model. Camera information is derived from the horizontal field of view to build the projection transform matrix, and the node outputs a positive conditioning pair (image-derived features plus projection data) and a negative conditioning pair (zeroed feature tensors) for classifier-free guidance.

Inputs

Parameter	Description	Data Type	Required	Range
`clip_vision_model`	DINOv3 ViT-L/16 ClipVision.	CLIP_VISION	Yes	—
`image`	Preprocessed image from ImageCropToMask (pad_factor=1.1 for Pixal3D).	IMAGE	Yes	—
`camera_angle_x`	Horizontal FOV in degrees (displayed as `fov`). Wire a MoGeGeometryToFOV (axis=‘horizontal’, unit=‘degrees’) for a per-image FoV (matches upstream default). Default: 49.13.	FLOAT	Yes	1.0 – 170.0 (step 0.01)

Outputs

Output Name	Description	Data Type
`positive`	The positive conditioning output containing the image-derived feature maps and projection data for Trellis2 generation.	CONDITIONING
`negative`	The negative conditioning output with zeroed feature tensors, used for classifier-free guidance.	CONDITIONING

Note: The camera_angle_x value is converted from degrees to radians internally, and the camera distance is computed from it to build the projection transform matrix. When the supplied vision model includes a NAF component, the node also produces high-resolution feature maps for the shape and texture stages.

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): 88e82b48fbe297c8e32ddd1b6659f196bda6f77fd53480bc021e85170d1923c7

​Inputs

​Outputs

Inputs

Outputs