ComfyUI nodes for video object segmentation using [a/SAMURAI](https://github.com/yangchris11/samurai) model.

SAMURAI Nodes for ComfyUI

# SAMURAI Points Input Node
    
This node allows you to select points of interest in the first frame of a video sequence for object segmentation.

## Inputs:
- **image**: Input video frames sequence (required)
- **start_frame**: Starting frame number for processing (default: 0)
- **refresh_input**: Trigger to refresh points selection (default: 0)

## Outputs:
- **POINTS**: Selected point coordinates for segmentation
- **LABELS**: Point labels (positive/negative)
- **START_FRAME**: Frame number to start processing from

## Usage:
1. Connect video input
2. Set starting frame number
3. Left click to add positive points (object)
4. Right click to add negative points (background)
5. Points and their labels will be passed to SAMURAI Refine node

## Note:
- Positive points (left click) indicate areas that belong to the object
- Negative points (right click) indicate areas that belong to the background

ComfyUI Node: SAMURAI Points Input

Category

Inputs

Outputs

Extension: SAMURAI Nodes for ComfyUI

More nodes in SAMURAI Nodes for ComfyUI

Run ComfyUI workflows without the setup