ComfyUI Node
Video Frame Extractor & Mask Generator 🎞️🅢🅜
Extracts/places frames from input/depth video into a new guidance video and generates corresponding masks. frame_selection_string takes precedence over depth frames.
Video Frame Extractor & Mask Generator 🎞️🅢🅜
- input_video_frames
- depth_video_frames
- master_inpaint_mask
- guidance_video_frames
- guidance_frame_masks
◄total_output_frames81►
◄frame_selection_string0, 10:20►
◄empty_frame_fill_level0.50►
CategorySteerable-Motion
Inputs (6)
| Name | Type | Default | Description |
|---|---|---|---|
| input_video_frames | IMAGE | Input video frames (IMAGE batch) to extract from. | |
| total_output_frames | INT | 811–10000 | Total number of frames for the output guidance video and masks. Must satisfy: (frames - 1) divisible by 4. |
| frame_selection_string | STRING | 0, 10:20 | Comma-separated integers or ranges (e.g., 0, 5, 10:15, 20) of frames to extract from input video. Takes precedence over depth_frames. |
| empty_frame_fill_level | FLOAT | 0.500–1 | Grayscale level (0.0 black, 1.0 white) for frames not explicitly selected or filled by depth. |
| depth_video_framesopt | IMAGE | Optional depth frames (IMAGE batch). Placed if the slot is not already filled by frame_selection_string. | |
| master_inpaint_maskopt | MASK | Optional master inpaint mask. If provided, it defines the entire output mask, overriding masks for selected/depth frames. |
Outputs (2)
| Name | Type | Description |
|---|---|---|
| guidance_video_frames | IMAGE | — |
| guidance_frame_masks | MASK | — |