ComfyUI Node
Grounding Mask Detector
A ComfyUI node in grounding with 9 inputs and 3 outputs.
Grounding Mask Detector
- model
- image
- masks
- overlaid_mask
- text
◄promptSegment the main object in the image►
◄confidence_threshold0.30►
◄florence2_max_tokens1024►
◄florence2_num_beams3►
◄sa2va_max_tokens2048►
◄sa2va_num_beams1►
◄seed42►
Categorygrounding
Inputs (9)
| Name | Type | Default | Description |
|---|---|---|---|
| model | MASK_MODEL | — | |
| image | IMAGE | — | |
| prompt | STRING | Segment the main object in the image | For Florence-2: descriptive phrase. For SA2VA: explicit segmentation instruction (e.g., 'Segment the person') |
| confidence_threshold | FLOAT | 0.300–1 | Confidence threshold for mask filtering (where applicable) |
| florence2_max_tokensopt | INT | 1024256–4096 | [Florence-2 only] Maximum tokens for generation |
| florence2_num_beamsopt | INT | 31–5 | [Florence-2 only] Beam search width |
| sa2va_max_tokensopt | INT | 2048512–8192 | [SA2VA only] Maximum tokens for generation |
| sa2va_num_beamsopt | INT | 11–5 | [SA2VA only] Beam search width |
| seedopt | INT | 420–4294967295 | Fixed seed for reproducible results |
Outputs (3)
| Name | Type | Description |
|---|---|---|
| masks | MASK | — |
| overlaid_mask | IMAGE | — |
| text | STRING | — |