September 1, 2026AIExplore

How to Turn Images into Videos with Luma

Animate approved stills in Dream Machine with motion prompts that respect lighting, composition, and product shape from the reference image.

Luma

Image to video in Luma Dream Machine animates an uploaded still while inheriting lighting, color, and composition from the reference. Prompt motion and camera behavior carefully when brand approval matters. Official Ray3 materials note improved image to video fidelity. Credit cost depends on model, resolution, and duration. Confirm details on lumalabs.ai/pricing. Overview: /explore/luma.

This guide covers the workflow in detail. Related Luma articles: /blog/how-to-use-luma-dream-machine-for-text-to-video, /blog/how-to-edit-and-refine-videos-iteratively-in-luma, /blog/how-to-use-keyframes-and-loops-in-luma.

When this workflow is the right job

Use this when photography or design stills are already approved. Prefer text to video when inventing a scene. Prefer keyframes when you need a defined start and end still.

Step by step workflow

1. Choose an approved still

Prefer clean product edges, readable labels where needed, and lighting you want to keep.

Scenario:
A realistic luma workflow is being prepared for Use Luma Dream Machine for Text-to-Video using luma. The starting requirement is: Still QA [ ] Product shape clear [ ] Label readable if required [ ] Background clean enough to hold [ ] Aspect matches delivery need. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Still QA [ ] Product shape clear [ ] Label readable if required [ ] Background clean enough to hold [ ] Aspect matches delivery need" that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Inputs:
Use the actual source material required for this task, together with the intended audience, destination, format, and quality requirements. Specify relevant files, text, URLs, structured data, reference assets, dimensions, language, tone, timing, model or generation settings, credentials, field mappings, or integration values when they apply. Keep secrets out of the example and use the product's supported credential or configuration mechanism.

Workflow:
1. Start with one representative input that contains the important characteristics of the real workload.
2. Configure luma specifically for the requested task and make the important settings explicit.
3. Run the first pass and inspect the result against the objective.
4. Correct only the settings or source material responsible for a failed requirement instead of changing everything at once.
5. Validate the intermediate result before passing it to the next step.
6. When the result meets the requirements, save the successful configuration so the same process can be repeated consistently.
7. If this is a multi-step workflow, verify each handoff and keep a clear fallback or review path for failures.

Requirements:
The example must use supported functionality only. Do not invent product features, pricing, limits, integrations, model names, API behavior, or unavailable settings. Preserve important source data and formatting. Validate required fields before processing, handle empty or invalid input explicitly, prevent accidental duplicate processing where relevant, and stop or route the item for review when a required step fails. For credentials, use the supported secure configuration rather than placing secrets directly in the content.

Expected output:
The final result should directly satisfy the "Still QA [ ] Product shape clear [ ] Label readable if required [ ] Background clean enough to hold [ ] Aspect matches delivery need" requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

2. Prompt motion only

Describe what moves. Avoid inventing new objects that contradict the still.

Scenario:
A realistic luma workflow is being prepared for Use Luma Dream Machine for Text-to-Video using luma. The starting requirement is: Product slowly rotates revealing all angles, soft studio lighting, minimal camera movement, premium commercial aesthetic, product details stay in focus, clean background.. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Product slowly rotates revealing all angles, soft studio lighting, minimal camera movement, premium commercial aesthetic, product details stay in focus, clean background." that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Product slowly rotates revealing all angles, soft studio lighting, minimal camera movement, premium commercial aesthetic, product details stay in focus, clean background." requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

3. Generate and inspect

Watch for warping on thin geometry, label drift, and unwanted background invention.

Scenario:
A realistic luma workflow is being prepared for Use Luma Dream Machine for Text-to-Video using luma. The starting requirement is: Inspect [ ] Product silhouette stable [ ] No melted edges [ ] Lighting still matches still [ ] Motion amount feels intentional. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Inspect [ ] Product silhouette stable [ ] No melted edges [ ] Lighting still matches still [ ] Motion amount feels intentional" that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Inspect [ ] Product silhouette stable [ ] No melted edges [ ] Lighting still matches still [ ] Motion amount feels intentional" requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

4. Tighten or Modify

Reduce motion, lock camera, or Modify small lighting changes instead of regenerating from a new still.

Scenario:
A realistic luma workflow is being prepared for Use Luma Dream Machine for Text-to-Video using luma. The starting requirement is: Reduce motion to a subtle slow orbit. Keep product centered. Static camera if warp appears.. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Reduce motion to a subtle slow orbit. Keep product centered. Static camera if warp appears." that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Reduce motion to a subtle slow orbit. Keep product centered. Static camera if warp appears." requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Practical image to video prompts

Product rotate

Subtle steam

Scenario:
A realistic luma workflow is being prepared for Use Luma Dream Machine for Text-to-Video using luma. The starting requirement is: Gentle steam rises from the cup in the still, camera locked, soft morning light preserved, no new props.. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Gentle steam rises from the cup in the still, camera locked, soft morning light preserved, no new props." that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Gentle steam rises from the cup in the still, camera locked, soft morning light preserved, no new props." requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Slow orbit sneaker

Scenario:
A realistic luma workflow is being prepared for Use Luma Dream Machine for Text-to-Video using luma. The starting requirement is: Slow orbit around the sneaker, product stays centered, soft daylight from the still, shallow depth of field, no background change.. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Slow orbit around the sneaker, product stays centered, soft daylight from the still, shallow depth of field, no background change." that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Slow orbit around the sneaker, product stays centered, soft daylight from the still, shallow depth of field, no background change." requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Fabric settle

Scenario:
A realistic luma workflow is being prepared for Use Luma Dream Machine for Text-to-Video using luma. The starting requirement is: Fabric settles softly in a gentle breeze, camera static, preserve material texture and color from the reference.. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Fabric settles softly in a gentle breeze, camera static, preserve material texture and color from the reference." that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Fabric settles softly in a gentle breeze, camera static, preserve material texture and color from the reference." requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Bottle turntable

Scenario:
A realistic luma workflow is being prepared for Use Luma Dream Machine for Text-to-Video using luma. The starting requirement is: Turntable style slow rotation of the bottle, studio lighting locked, label readable, clean seamless backdrop.. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Turntable style slow rotation of the bottle, studio lighting locked, label readable, clean seamless backdrop." that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Turntable style slow rotation of the bottle, studio lighting locked, label readable, clean seamless backdrop." requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Plant leaf sway

Scenario:
A realistic luma workflow is being prepared for Use Luma Dream Machine for Text-to-Video using luma. The starting requirement is: Leaves sway gently in a soft breeze, camera locked, preserve pot and table composition from the still.. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Leaves sway gently in a soft breeze, camera locked, preserve pot and table composition from the still." that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Leaves sway gently in a soft breeze, camera locked, preserve pot and table composition from the still." requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Car exterior glide

Scenario:
A realistic luma workflow is being prepared for Use Luma Dream Machine for Text-to-Video using luma. The starting requirement is: Camera slowly glides along the car side, reflections move naturally, keep body color and trim from the still, restrained motion.. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Camera slowly glides along the car side, reflections move naturally, keep body color and trim from the still, restrained motion." that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Camera slowly glides along the car side, reflections move naturally, keep body color and trim from the still, restrained motion." requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Jewelry sparkle

Scenario:
A realistic luma workflow is being prepared for Use Luma Dream Machine for Text-to-Video using luma. The starting requirement is: Subtle light sparkle across metal surfaces, micro camera push, keep gem placement exact, dark velvet background unchanged.. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Subtle light sparkle across metal surfaces, micro camera push, keep gem placement exact, dark velvet background unchanged." that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Subtle light sparkle across metal surfaces, micro camera push, keep gem placement exact, dark velvet background unchanged." requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Food steam plate

Scenario:
A realistic luma workflow is being prepared for Use Luma Dream Machine for Text-to-Video using luma. The starting requirement is: Soft steam rises from the plated dish, locked camera, preserve garnish placement, warm restaurant lighting from the still.. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Soft steam rises from the plated dish, locked camera, preserve garnish placement, warm restaurant lighting from the still." that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Soft steam rises from the plated dish, locked camera, preserve garnish placement, warm restaurant lighting from the still." requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Book page flutter

Scenario:
A realistic luma workflow is being prepared for Use Luma Dream Machine for Text-to-Video using luma. The starting requirement is: Top page flutters lightly, camera static, keep cover art and desk layout from the reference.. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Top page flutters lightly, camera static, keep cover art and desk layout from the reference." that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Top page flutters lightly, camera static, keep cover art and desk layout from the reference." requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Candle flame

Scenario:
A realistic luma workflow is being prepared for Use Luma Dream Machine for Text-to-Video using luma. The starting requirement is: Candle flame flickers gently, locked frame, warm practical light preserved, no new objects on the table.. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Candle flame flickers gently, locked frame, warm practical light preserved, no new objects on the table." that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Candle flame flickers gently, locked frame, warm practical light preserved, no new objects on the table." requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Watch second hand

Scenario:
A realistic luma workflow is being prepared for Use Luma Dream Machine for Text-to-Video using luma. The starting requirement is: Second hand moves smoothly, macro lock off, keep dial typography and metal finish from the still.. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Second hand moves smoothly, macro lock off, keep dial typography and metal finish from the still." that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Second hand moves smoothly, macro lock off, keep dial typography and metal finish from the still." requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Chair product

Scenario:
A realistic luma workflow is being prepared for Use Luma Dream Machine for Text-to-Video using luma. The starting requirement is: Slow push in on the chair, preserve wood grain and fabric, soft showroom light, no room redesign.. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Slow push in on the chair, preserve wood grain and fabric, soft showroom light, no room redesign." that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Slow push in on the chair, preserve wood grain and fabric, soft showroom light, no room redesign." requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Perfume mist

Scenario:
A realistic luma workflow is being prepared for Use Luma Dream Machine for Text-to-Video using luma. The starting requirement is: Soft mist drifts past the bottle, static camera, keep silhouette and label, elegant commercial mood.. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Soft mist drifts past the bottle, static camera, keep silhouette and label, elegant commercial mood." that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Soft mist drifts past the bottle, static camera, keep silhouette and label, elegant commercial mood." requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Shoe lace settle

Scenario:
A realistic luma workflow is being prepared for Use Luma Dream Machine for Text-to-Video using luma. The starting requirement is: Laces settle after a gentle tug, minimal camera move, preserve sole and upper materials.. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Laces settle after a gentle tug, minimal camera move, preserve sole and upper materials." that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Laces settle after a gentle tug, minimal camera move, preserve sole and upper materials." requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Laptop screen glow

Scenario:
A realistic luma workflow is being prepared for Use Luma Dream Machine for Text-to-Video using luma. The starting requirement is: Screen glow pulses softly, locked camera, keep desk props and keyboard layout from the still.. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Screen glow pulses softly, locked camera, keep desk props and keyboard layout from the still." that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Screen glow pulses softly, locked camera, keep desk props and keyboard layout from the still." requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Flower vase

Scenario:
A realistic luma workflow is being prepared for Use Luma Dream Machine for Text-to-Video using luma. The starting requirement is: Petals tremble in a soft breeze, camera locked, preserve vase shape and water line.. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Petals tremble in a soft breeze, camera locked, preserve vase shape and water line." that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Petals tremble in a soft breeze, camera locked, preserve vase shape and water line." requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

How to improve image to video results

Treat the still as law for composition. Prompt should describe motion, not a new scene.

Heavy camera moves on thin product edges often warp. Prefer subtle orbit or locked camera.

If labels must stay perfect, plan overlays in post and keep motion gentle.

Reduce motion amplitude before changing the still when warping appears.

Match delivery aspect ratio with the source crop so the product does not get reframed oddly.

Use Modify for color temperature tweaks after a good motion pass rather than regenerating motion.

Keep background language minimal: clean background or keep backdrop often outperforms invent a new set.

For multi SKU work, reuse the same motion template and swap stills.

Inspect hands or reflections carefully when they appear in lifestyle stills.

Prefer approved photography over generated stills when legal review cares about product truth.

If the model invents props, add an explicit ban list in the next prompt.

Credit costs scale with model and resolution. Draft lower cost passes before finals.

Prompting guidance that holds up in review

Lead with the verb of motion: rotates, sways, settles, rises, glides.

Say what must not change: product shape, label placement, palette, backdrop.

Prefer minimal camera movement language for packaging and jewelry.

Describe light as preserved from the still when brand lighting is already approved.

Keep prompts shorter than text to video briefs because the image already carries look.

Limitations to respect

Image to video still spends credits by model and resolution. Confirm the live table on lumalabs.ai/pricing.

Thin geometry and small type are common failure points. Plan post for critical marks.

Large invented camera moves can break product truth even when the still was perfect.

Extend only after the first seconds look brand safe.

Field tips from real boards

Combine with Extend only after the first few seconds look brand safe.

Keyframes help when you need a second approved end still. Plain image to video is enough for simple motion.

Document which motion verbs work for each category: rotate for bottles, sway for plants, settle for fabric.

Stakeholder reviews should watch full plays, not thumbnails, because warp shows in motion.

When a still is already approved, treat motion amplitude as the main risk dial. Start subtle, then increase only if edges stay clean.

Label drift often appears after the first full rotation. Catch it early by scrubbing the midpoint before you spend a final resolution pass.

Keep a motion verb library per category so SKUs share cadence even when photography differs.

If reflections invent logos, reduce camera travel and composite critical marks in post.

Crop the source still to the delivery aspect before generating so the product does not get reframed mid motion.

Lifestyle stills with hands need a tighter motion budget than clean pack shots.

After a good motion pass, use Modify for grade only. Do not reopen composition unless the silhouette failed.

When a still is already approved, treat motion amplitude as the main risk dial.

Label drift often appears after the first full rotation. Scrub the midpoint early.

Keep a motion verb library per category so SKUs share cadence.

If reflections invent logos, reduce camera travel and composite marks in post.

Crop the source still to delivery aspect before generating.

Lifestyle stills with hands need a tighter motion budget than pack shots.

After a good motion pass, use Modify for grade only.

Pack shot stills with clean edges tolerate more orbit than lifestyle frames with fingers near the product.

If the model invents a second product in the background, ban extra objects explicitly on the next pass.

Match the still white balance language in the prompt so motion does not imply a new lighting setup.

For transparent bottles, reduce motion first. Thin glass edges fail before opaque packaging does.

Save a motion recipe card per category: rotate, sway, settle, steam, glow. Reuse the verb, swap the still.

Client review should include a full playthrough at 100 percent zoom on edges, not only a social crop.

When the still has busy type, keep the camera locked and animate only secondary elements like steam.

Common mistakes

Asking for new objects that contradict the still
Heavy camera moves on thin product geometry
Ignoring label or logo drift
Skipping post for mandatory brand typography
Changing still and motion recipe at the same time
Using text to video when an approved still already exists

Related Luma articles: /blog/how-to-use-luma-dream-machine-for-text-to-video, /blog/how-to-edit-and-refine-videos-iteratively-in-luma, /blog/how-to-use-keyframes-and-loops-in-luma. Return to /explore/luma for the broader guide set.