Nodes/comfyui_cv/* → CV Array
ComfyUI Node

* → CV Array

Converts an IMAGE, MASK, LATENT or NPARRAY to a raw OpenCV ndarray (NPARRAY) for the low-level cv2.* nodes, auto-detecting the input type. Frame 0 of a batch is taken. IMAGE becomes uint8 BGR [H,W,3]; MASK becomes uint8 [H,W]; LATENT becomes float32 [H,W,C] with values untouched; an NPARRAY passes through unchanged. Useful in subgraphs whose boundary socket must accept an arbitrary type but process it as a CV array. For extra control (color_format / dtype / batch_index) use the dedicated 'Image → CV Array' / 'Mask → CV Array' / 'Latent → CV Array' nodes instead.

By bmad4ever·Created 3 months ago·Updated 2 days ago· 0
* → CV Array
  • input
  • nparray
Categoryimage/CV/low-level

Inputs (1)

NameTypeDefaultDescription
inputNPARRAY,IMAGE,MASK,LATENTIMAGE, MASK, LATENT or NPARRAY to convert to a raw ndarray. A LATENT link is processed in latent space: frame 0 becomes a float32 [H,W,C] array (any channel count), values untouched. Arithmetic ops (add, multiply, etc.) also accept a full LATENT batch ({samples: [B,C,H,W]}) — the whole batch flows through when both inputs have the same batch size. Accepts a ComfyUI IMAGE/MASK directly (frame 0 of a batch) or an NPARRAY. Arithmetic ops (add, multiply, etc.) process the full IMAGE batch when both inputs have the same batch size.

Outputs (1)

NameTypeDescription
nparrayNPARRAYRaw ndarray: uint8 BGR [H,W,3] from IMAGE, uint8 [H,W] from MASK, float32 [H,W,C] from LATENT, or the NPARRAY as-is.