ComfyUI Node
BFS Shot Noise
Optional structured starting noise. With every amount at 0 it is exactly RandomNoise. shared_base: the coarse noise pattern is the same for every shot / size / length (from base_seed), the detail per shot. static_share: a part identical in all frames. guide_structure: the guide's latent mixed in outside the mask. ref_tint: lean toward the reference's colours inside the mask. Experimental: the model was trained on plain noise, so use small amounts and compare with 0.
BFS Shot Noise
- shot
- vae
- guide_frames
- ref_image
- mask
- noise
- seed
◄seed0►
◄shot_seedsame for every shot►
◄shared_base0.00►
◄static_share0.00►
◄guide_structure0.00►
◄distributiongaussian►
◄spectrum_tilt0.00►
◄pyramid0.00►
◄ref_tint0.00►
◄base_seed-1►
◄overrides►
◄base_grid6►
◄time_block2►
CategoryBFS/shot loop
Inputs (18)
| Name | Type | Default | Description |
|---|---|---|---|
| seed | INT | 00–18446744073709550000 | The seed (the base seed when a shot is connected). |
| shot_seed | COMBO | same for every shot | Only with a shot connected: the seed of each shot (same / counting up / unrelated but repeatable). |
| shared_base | FLOAT | 0.000–1 | 0 = off (plain noise). The coarse pattern of the noise comes from base_seed instead of the shot's seed: every shot, any length or size, agrees on it; the fine detail stays per shot. 1 = fully shared. Try 0.5 - 1. |
| static_share | FLOAT | 0.000–0.9 | 0 = off. A part of the noise that is identical in every frame (temporally correlated noise). Try 0.1 - 0.3; too much freezes the motion. |
| guide_structure | FLOAT | 0.000–0.6 | 0 = off. Needs guide_frames + vae. Outside the mask the guide's latent is mixed into the noise, so the background starts closer to the guide (without a mask: everywhere, which also pulls toward the original person). Try 0.05 - 0.2. |
| distribution | COMBO | gaussian | gaussian = what the model was trained on. laplacian / student-t: more extreme values (stronger, sparser kicks). uniform: none. All at unit variance. Like AdvancedNoise's types, but for H3's video + audio latent (the audio noise always stays gaussian). |
| spectrum_tilt | FLOAT | 0.00-2–2 | 0 = off (white noise). > 0: more energy in the large blobs (fractal / pink noise, AdvancedNoise's 'fractal' alpha): the coarse layout, where the swap is decided, gets a harder push. < 0: more fine grain. Try 0.2 - 0.6; large values burn colours. |
| pyramid | FLOAT | 0.000–0.95 | 0 = off. Adds coarser layers of noise (each half the size, weight = this value per level): another way to put more energy in the large scales. Try 0.3 - 0.6. |
| ref_tint | FLOAT | 0.000–0.6 | 0 = off. Needs ref_image + vae. Inside the mask (everywhere without one) the noise leans toward the reference's average latent, i.e. its colours. Try 0.05 - 0.15. |
| shotopt | BFS_SHOT | From BFS Shot Planner / Unpack: per-shot seeds, and the shared base is read at the shot's position on the timeline. | |
| base_seedopt | INT | -1-1–18446744073709550000 | Seed of the shared coarse pattern and of static_share. -1 = the seed above. Put here the seed of a result that worked to keep its coarse noise while the detail changes with the seed. |
| overridesopt | STRING | Seeds for single shots: 'shot:seed, shot:seed' (shots count from 1). | |
| base_gridopt | INT | 61–32 | Blocks of the coarse pattern along the short side (the long side in proportion). Fewer = only the broadest layout is shared. |
| time_blockopt | INT | 21–32 | Latent frames per block of the coarse pattern. |
| vaeopt | VAE | The video VAE, for guide_structure and ref_tint. | |
| guide_framesopt | IMAGE | The guide video frames (guide_structure). | |
| ref_imageopt | IMAGE | The reference picture (ref_tint). | |
| maskopt | MASK | The person, one mask per frame (white = the person): guide_structure stays outside it, ref_tint inside it. |
Outputs (2)
| Name | Type | Description |
|---|---|---|
| noise | NOISE | For SamplerCustomAdvanced. |
| seed | INT | The seed this shot used. |