← Back to films

MULTIMODAL / PRODUCTION METHODS

Multimodal: from references to usable shots.

Product films, brand stories, shot generation and live-action compositing. Methods and tools, separate from the work itself.

OPERATOR NOTES / METHODS & IMPLEMENTATION

Generate shot by shot—or preserve the live-action plate.

When suitable footage is missing, lock product, identity and scene references, approve stills, then generate motion shot by shot. When a good live-action plate exists, preserve the background and performance and replace only what is needed. Finish with continuity, sound and pacing.

01What references are locked

Bind product, person, wardrobe and scene references separately. Unseen product details remain unknown; identity follows approved face and body references.

02When to generate motion

Define purpose, entry state, action, exit state and camera per shot. Approve the still first; repair only affected shots and dependencies.

03Why not regenerate everything

For a good live-action base with a local problem, use masks, tracking, shadows and reflections. Full regeneration can damage what already works.

04How the film is accepted

Check product and action per shot, then review rhythm, sound and subtitles across the whole film. Decodable, ready-to-edit and approved are distinct states.

WORKFLOWBrief & true referencesShot cardsStill approvalGenerate / CompositeSound & editFull-film review

DECOMPOSITION / CONDITIONING / EVALUATION

What happens behind each shot.

Decompose the reference into timed shots, bind product and identity sources, then choose a generation route. Each step defines a decision, deliverable and response to failure.

Turn a reference into reproducible shot decisions.

Locate candidate cuts, align speech and visible text, then identify what each shot does: hook, demonstration, evidence or objection handling. Transfer the communication pattern, not unverified product facts.

StepDecision and method
TimingFrame differences locate candidates. Pans, flashes and dissolves need review; inspect both sides of boundaries and action transitions.
Speech and textASR / OCR produce timed observations. Verify names, numbers and units before treating them as product evidence.
Transfer the DNAAn opening-and-pouring action does not fit a device that never opens. Retain the operation–change–verification structure instead.
DeliverableShot ID, in/out time, purpose, action, camera, sound, keep/replace decisions and the exact source moment.
ANALYZEDecode & cut candidatesTimed observationsShot purposeKeep / replace map
PRODUCERole-scoped referencesApproved keyframeSupported backendActual take review
REPAIRLocate defectDependency closureBudget decisionRecheck & edit

RUNNABLE / ACTUAL VIDEO INPUT

Video analysis: locate moments to review.

A local inspection of the first TikTok DNA recreation video scanned the first 60 seconds and decoded 1,438 frames. These are pixel-change candidates, not automatically recognized story beats or a quality score. The source was not modified.

Observed outputHow to use it
15.070 s / 30.140 s / 45.210 sThree scene-change candidates. Review neighboring frames to distinguish cuts, transitions and fast movement.
Low-change intervals: 0No hits at these settings does not establish the absence of deformation, flicker, still shots or other errors.
Last observed frame: 59.960 sThe last frame of this scan, not the full source duration or final-film acceptance.
Parameters & provenancescene_threshold=10; freeze_seconds=1; freeze_noise=0.001; max analysis width640px. Output includes FFmpeg version and source SHA-256.
PIXEL EVENTS / PYTHON + FFMPEG
python3 scripts/multimodal.py inspect /absolute/path/video.mp4 \
  --seconds 60 --scene-threshold 10 --freeze-seconds 1

The tool uses FFmpeg scene-change and low-change detection. It does not analyze audio, recognize identity/products or explain virality. Thresholds are configurable; candidates require source review.

WORKED EXAMPLE / DEPENDENCY-AWARE REPAIR

Why changing shot two need not rebuild the film.

Synthetic example: S3 consumes S2's exit and S4 consumes S3's; S1 and S5 are independent. Changing S2 flags S2/S3/S4 for review while preserving S1/S5. Edges represent actual render references, not playback order.

RECHECKS2 / changedS3 / exit-dependentS4 / exit-dependent
PRESERVES1 / independentS5 / independent
ConditionCalculation and decision
2 units per attempt; 6 remainingThree affected shots × one attempt =6. READY_FOR_REVIEW means review the plan, not approval or execution.
Only5 remaining or an exhausted shotNEEDS_PRODUCTION_DECISION: choose scope, reshoot or budget; no automatic retry.
A descendant is flaggedCheck whether its consumed reference really changed. Reuse unaffected takes and inspect edit junctions separately.
DEPENDENCY PLAN / NO GENERATION CALLS
python3 scripts/multimodal.py plan /absolute/path/repair.json
# changed: [S2]
# recheck_or_rebuild: [S2, S3, S4]
# unaffected_by_declared_changes: [S1, S5]
# one_attempt_estimate: 6

IMPLEMENTED / LOCAL / OPTIONAL

What runs today—and what needs a model.

CapabilityCurrent statusBoundary
Pixel-event inspection + repair planningPublic and runnable in this updatePixels produce timed candidates; task edges produce repair scope. Neither certifies identity or SKU.
Product ROI / source hashes / reference roles / checkpointsExisting local brand-film implementationROI crops approved coordinates; hashes prevent stale-source reuse, not visual mistakes.
Continuity exit selectionExisting local implementationThree tail candidates ranked by motion and blur; not semantic action understanding. A fallback is not an optimal frame.
Video request compilationExisting local adapterAn adapter is neither universal model support nor proof of a live integration or completed render.
ASR/OCR, subject features, optical flow, SAM 2 / VBenchOptional integrations; not bundled or run hereCalibrate on project samples before assisting review; these are not claimed as self-trained foundation models.
Technical sources

FFmpeg powers the actual media inspection. PySceneDetect is an optional shot detector, SAM 2 provides segmentation/propagation, and VBench informs multidimensional evaluation—not guaranteed automatic production.

Implementation, inputs and local checks

Use jingqiu-DTC for short product storyboards and the existing brand-film method for continuity. The shots check organizes timed review/repair records; it does not read pixels. Reframe's default 9:16 and voice replacement are unsuitable for preserving landscape footage and original sound.

LOCAL REVIEW / NO ACCOUNT WRITES
python3 scripts/review.py examples.json --example shots

This runs synthetic input checks, not a live account integration or an experiment proving lift.

Generation, visual approval and advertising performance are separate checks. Verify product and identity fidelity, review continuity and sound, then assess effectiveness using real campaign data.