7da6f7427f
Shot Planner: up to 12 reference pictures, @Picture chips, {details@Picture n} (1.113.0)
alisson-anjosandClaude Opus 5.52026-10-18 00:18:36 +00:00
084e1d22b5
Shot loop: a new LoRA, seed or sampler runs the shots again; no preview on BFS Shot Join (1.112.0)
alisson-anjosandClaude Opus 5.52026-10-17 14:48:59 +00:00
55b120d14b
Video Compare: choose the format of save / rec: MP4, small MP4, GIF or the browser file (1.111.0)
alisson-anjosandClaude Opus 5.52026-10-17 12:34:44 +00:00
526745455f
Source blur: choose what stays sharp, masked area, Sapiens2 segmenter, invert colours (1.110.0)
alisson-anjosandClaude Opus 5.52026-10-17 12:02:24 +00:00
78a806103e
Video Compare: save / record, reference picture, side-by-side size fix; wheel zoom over panels (1.109.0)
alisson-anjosandClaude Opus 5.52026-10-17 11:50:49 +00:00
ce3ca59b17
Shot loop: run again without changes only assembles; unload models before the join (1.108.0)
alisson-anjosandClaude Opus 5.52026-10-16 22:15:41 +00:00
187c9a1776
Find people / Cast without insightface: OpenCV Zoo YuNet + SFace (Apache-2.0, run by OpenCV, downloaded once into models/opencv_face); InsightFace only when installed; SAM 3.1 + colours as last resort
alisson-anjosandClaude Opus 5.52026-10-09 19:00:36 +00:00
a04f3116cf
{details} filled automatically by the render nodes too: the planner leaves it when it has no VLM, BFS Shot H3 Conditioning / Shot H3 Duet (new vlm input) describe the shot's references
alisson-anjosandClaude Opus 5.52026-10-09 12:48:30 +00:00
84a953e8c5
source_blur (SAM 3.1, keeps the mouth): blur the source person in the guide / native ref video / duet panel so the model reads pose, motion and lips but has no face to copy - Shot Conditioning, Shot Duet, Duet, Duet Conditioning
alisson-anjosandClaude Opus 5.52026-10-09 12:10:27 +00:00
0bf03a8fb7
Planner: chunked upload for big files (8 MB pieces to /bfs/upload/chunk, assembled on disk) - any size, past ComfyUI's 100 MB --max-upload-size
alisson-anjosandClaude Opus 5.52026-10-09 09:03:16 +00:00
390f9e8714
Planner UI: ✕ clear all - removes every reference from the plan (global, shots, cast) and the recent list, after a confirm
alisson-anjosandClaude Opus 5.52026-10-09 07:50:53 +00:00
515ab73c70
Planner UI: collapsible sections keep what the user clicked (re-renders snapped them shut); ✕ clear for the recent reference pictures
alisson-anjosandClaude Opus 5.52026-10-09 07:47:46 +00:00
2e64d989f6
BFS Shot H3 Duet: continuity 'history' too (the previous end before frame 0, panel strip filled with the source under it); shared shot_history() pops it from the shot
alisson-anjosandClaude Opus 5.52026-10-09 02:17:41 +00:00
b1e0fcbbb6
Shot Planner: continuity 'history' is the default for new plans and unset shots (cuts still skip it)
alisson-anjosandClaude Opus 5.52026-10-09 02:14:20 +00:00
0b5315c45a
History / chain: load the previous result on the CPU, slice uint8 before converting, keep history as uint8 in the shot and drop it after conditioning (no idle RAM/VRAM)
alisson-anjosandClaude Opus 5.52026-10-09 01:39:46 +00:00
a03e0be342
Shot Planner: continuity 'history' - the last 17/34/51 frames of the previous result placed just before frame 0 on H3's timeline (sliding history, after Akatz's H3 Relay / Ethanfel's Context Loop), with every mask mode, crop and the duet canvas; cuts reset it
alisson-anjosandClaude Opus 5.52026-10-09 01:37:36 +00:00
78aa214584
Shot Planner: {details} described automatically before any SAM 3 work (it ran after the masks, which breaks the VLM: the button worked, the run did not); never leaves a hole in the sentence
alisson-anjosandClaude Opus 5.52026-10-08 23:03:47 +00:00
596a1c2538
Duet prompts: TSC's newer strategy - the finished split screen as plain footage (no kept footage / generated / in sync), the pinned region named by place + one clause of what is seen ({panel_region}/{video_region}: top quarter, bottom three quarters...), motion beat by beat with times on the video clock, sounds of the new scene
alisson-anjosandClaude Opus 5.52026-10-08 21:18:07 +00:00
4e4fc20fdd
Duet: optional VLM prompt - writes it from the task, or fills the {...} fields of a hand-written template (TSC's); setting picture described as <Subject 2>
alisson-anjosandClaude Opus 5.52026-10-08 21:08:35 +00:00
ad99a9f91a
Duet: panel_static (TV static over face/person, SAM 3) + setting picture input + BFS Setting Picture node
alisson-anjosandClaude Opus 5.52026-10-08 20:52:18 +00:00
f763b51779
BFS H3 Duet: side_by_side output - the same comparison as BFS Shot Join (original | references | result, info on top, prompt below) (1.83.0)
alisson-anjosandClaude Opus 5.52026-10-08 20:21:36 +00:00
2b462b639a
BFS H3 Duet: megapixels input removed again (the size is computed in the workflow instead) (1.82.1)
alisson-anjosandClaude Opus 5.52026-10-08 19:58:46 +00:00
864a28b161
BFS H3 Duet / Duet Conditioning: megapixels - output size from the clip's aspect ratio at that area (0 = width / height) (1.82.0)
alisson-anjosandClaude Opus 5.52026-10-08 19:54:02 +00:00
e43cc14d9a
Points: Track -> whole video - one selection tracked by SAM 3 through the whole video or range (chunked, forward and backward, re-found by text after cuts); shots without their own selection take their part (1.81.0)
alisson-anjosandClaude Opus 5.52026-10-08 19:47:21 +00:00
4fab22caa9
Shot Planner range: a player in the Video tab, start / end in frames (start here / end here take the player's exact frame; seconds of older plans still read), aligned layout (1.80.0)
alisson-anjosandClaude Opus 5.52026-10-08 19:09:13 +00:00
96f7e8a1b1
VLM descriptions: temperature and seed (0 = greedy as before), a dice button to try another description, the short preset capped at 30 words (1.79.1)
alisson-anjosandClaude Opus 5.52026-10-08 18:02:53 +00:00
34ffcef172
Shot Planner: Start at / End at range - only that part of the video is split, analysed, generated and joined; timeline dims the rest; max total seconds counts from the start (1.79.0)
alisson-anjosandClaude Opus 5.52026-10-08 17:42:42 +00:00
9ae5fc9614
Fit the prompt: only for the short preset, the VLM is told to reply with the blank's words only, and an echo of the prompt sentence is cut; Checks warn when {details} is used with a paragraph preset (1.78.2)
alisson-anjosandClaude Opus 5.52026-10-08 15:14:58 +00:00
7d5f60ce86
insightface and onnxruntime are optional: no pin in the requirements (the exact insightface pin and its own requirements conflicted with many environments); Find people / LTX Identity say what to install (1.78.1)
alisson-anjosandClaude Opus 5.52026-10-08 13:50:54 +00:00
8e573275b8
Fit the prompt: descriptions ({details}, {target}) are written for the prompt sentence that holds them (blank in place, article matched); {target} is described automatically at run time for shots with a selection and no text (1.78.0)
alisson-anjosandClaude Opus 5.52026-10-08 13:34:08 +00:00
a5535dc77c
Describe refs for any subject (person, creature, robot, animal, cartoon: says what it is first); new default preset 'short' - a noun phrase that fits '<Subject 1> is the {details} shown in <Picture 1>'; {target} describes animals/creatures too (1.77.0)
alisson-anjosandClaude Opus 5.52026-10-08 11:35:11 +00:00
7af3fa5618
pose look: clearer error (package vs model, with the download link) and no restart needed after adding the model (1.76.1)
alisson-anjosandClaude Opus 5.52026-10-07 16:07:09 +00:00
39610ab1ef
mask_guide_look: what the mask guide shows of the subject - grey blurred (default), colour, silhouette, edges, or pose (people, YOLO pose model from models/ultralytics, no new packages) (1.76.0)
alisson-anjosandClaude Opus 5.52026-10-07 15:47:14 +00:00
aa92d72990
Faster masks: separable dilation / blur (same result, ~7x faster on few CPU threads) and the generation mask computed once per shot (1.75.1)
alisson-anjosandClaude Opus 5.52026-10-07 15:35:14 +00:00
f1388f0463
mask_guide (renamed from person_guide, works for any subject): masked region as an extra guide, instead of the full guide, or as a native reference video (<Video n>, with mask_ref_size); Mask only shots only; Frame + paste takes a per-shot 'new subject' text for the new outline (1.75.0)
alisson-anjosandClaude Opus 5.52026-10-07 14:51:56 +00:00
5c93d23c88
H3 Conditioning: experimental person_guide - a guide latent with only the masked person (rest grey), added next to the full guide or instead of it (1.74.0)
alisson-anjosandClaude Opus 5.52026-10-07 14:45:55 +00:00
7d9d7a9e26
Frame + paste mode: the whole frame is generated (pose follows the guide, the new person may be bigger) and BFS Shot Join pastes only the person (shot mask + the new outline found by SAM 3) onto the original frame; mask strength renamed mask opacity (1.73.0)
alisson-anjosandClaude Opus 5.52026-10-07 14:22:16 +00:00
1de7753be9
Mask strength per shot (mask modes): the masked rows start at strength x the noise level and keep part of the original, like an opacity (1.72.0)
alisson-anjosandClaude Opus 5.52026-10-07 14:09:17 +00:00
aa545097b7
External masks instead of SAM 3: a mask video per shot, the planner's optional mask input (whole video) or a mask video for the whole video in the panel; used by every mode, {target}, the setting picture and the comparison (1.71.0)
alisson-anjosandClaude Opus 5.52026-10-07 14:01:13 +00:00
1b04a822b3
Console logs for masks: H3 Conditioning prints each shot's mode, where its mask comes from, coverage (and empty frames) and how much of the latent is regenerated; SAM 3 runs print what, how many frames and how long (1.70.1)
alisson-anjosandClaude Opus 5.52026-10-07 13:45:07 +00:00
61a9e732f2
Shot Planner refactor: tabbed panel, per-shot generation modes, Copy to other shots, checks (1.70.0)
alisson-anjosandClaude Opus 5.52026-10-07 11:54:22 +00:00
736a7016e8
Shot Planner: wider node by default (1180 px, at least 1040 for older workflows), panel at least 900 px wide
alisson-anjosandClaude Opus 5.52026-10-07 11:32:25 +00:00
