How to Make an AI Character Hold a Product Without Changing Its Shape
Treat the product as a measured brand asset: lock silhouette, proportions, label placement, colors and logo. Avoid asking the model to invent unseen surfaces during aggressive motion when fidelity matters.
Treat the product as a measured brand asset: lock silhouette, proportions, label placement, colors and logo. Avoid asking the model to invent unseen surfaces during aggressive motion when fidelity matters.
What is actually failing
This is a product fidelity problem, not a signal that you need a longer prompt. Modern video models can follow references and motion instructions, but every new angle, occlusion, contact point or camera move introduces visual information the model must infer. The safest workflow is to identify which state must remain fixed and which state is allowed to change.
Hands create occlusion and contact deformation. Keep the grip simple, define which hand holds which part, and avoid hiding the product’s key silhouette.
Use a locked-state workflow
- Choose the last known-good state. Save the approved reference or clean frame before changing anything.
- Write down the invariants. Keep a short list of details that must not change; do not bury them inside prose.
- Change one production variable. Add the new action, angle, camera move or interaction without redesigning the subject.
- Reduce ambiguity. If the shot fails, simplify motion, shorten the clip, expose the subject more clearly or add the missing reference angle.
- Verify before extending. Only use a clip as the next reference after it passes a frame-by-frame continuity check.
A prompt structure that is easier to debug
LOCKED: - subject / product identity: [unchanged] - wardrobe / materials / colors: [unchanged] - location anchors: [unchanged] - lighting direction: [unchanged] CHANGE ONLY: - action: [...] - camera: [...] - start state: [...] - end state: [...] DO NOT INTRODUCE: - new wardrobe, props, cuts, camera angles or redesigns - changes to logo, facial identity, product proportions or scene layout
When another reference helps — and when it hurts
Keep motion complexity below the fidelity limit
Verification checklist
- Does the face or product silhouette remain stable at the hardest point of motion?
- Do hair, wardrobe, labels and accessories survive the whole clip?
- Are start position, end position and screen direction correct?
- Did the model add an unrequested cut, prop, person or camera move?
- Can the final clean frame safely become the starting reference for the next shot?
Where the AI Video Kit fits
Doing this more than once? Standardize it.
Once making an AI Character Hold a Product without Changing Its Shape becomes recurring work, a reusable process is faster than reconstructing the same checks from memory. The Brightory AI Video Consistency & Production Kit packages this kind of work into continuity sheets, shot-planning templates, reusable prompt blocks and production QA checks, so you can carry the method into the next project instead of starting over.
Product continuity benchmark
Treat the product as a controlled asset with silhouette, proportions, colors, label placement and material properties that must survive motion—not as a descriptive noun.
- Build a product lock sheet.
- Provide reference coverage for angles the shot reveals.
- Separate object interaction from camera motion.
- Protect hand/contact occlusions.
- QC the worst frame, not only the hero frame.
Acceptance test
Silhouette, branding placement, relative scale and key colors must remain visually consistent from the first usable frame through the hardest contact/motion frame to the last.
Decision rule
Review product continuity like packaging QA: identity is a set of visual properties, not a general impression.
Product QC matrix
| Check | Compare | Reject when |
|---|---|---|
| Silhouette | Contour and distinctive corners. | Geometry or aspect ratio changes. |
| Branding | Logo/label position and orientation. | Text bends, mirrors or migrates. |
| Scale | Object relative to hands/body/set. | Scale jumps during interaction. |
| Material/color | Hue, gloss and texture cues. | Identity changes beyond lighting explanation. |
Worked example: isolate a pickup failure
If a bottle label stretches when a hand picks it up, save the last clean frame and record silhouette, cap ratio, label position and key colors. First test the pickup with a static camera. If the product survives, add the intended camera move in a separate pass.
Inspect the most occluded frame and the first frame after the label becomes visible again. Those frames reveal whether the model reconstructed hidden geometry incorrectly. This is more diagnostic than adding “keep product perfectly consistent” to the prompt.
Operational notes for this workflow
Keep a clean product reference sheet separate from the cinematic references. Include the angles needed for the shot and record label/logo geometry, dominant colors and proportions. If hands or props cover the product, identify the frame of maximum occlusion and the first clean frame after reveal as mandatory QC points.
When a camera move and product interaction both occur, test the interaction with simpler framing first. This establishes whether the geometry failure comes from contact/occlusion or from the camera motion.
Final continuity checklist
- Known-good state: did this test begin from an approved reference, frame or clip?
- Single delta: can you name the one visual or motion variable that was allowed to change?
- Weakest frame: did you inspect the highest-motion, most occluded or most oblique frame rather than only the opening image?
- Boundary: does the shot connect cleanly to the state before and after it?
- Promotion rule: would you trust this exact result as the reference for the next generation?
If the answer to the final question is no, the clip has not passed continuity even if several frames look excellent. Repair the failed state before extending the sequence.
- Runway — AI Character References: Tips & Techniques
- Runway — Gen-4 Image References
- Google AI for Developers — Veo 3.1 reference images
- Adobe Firefly — Motion reference
- Adobe Firefly — Generate videos using images / first and last frames
AI video controls differ by model and can change quickly. The workflow below distinguishes durable production practice from model-specific features; re-check the linked official documentation before relying on a current limit or control.