Create a product color variation
Keep camera movement and product shape while testing a different material or color before approving a reshoot or render.
Video-to-video editing on IMG AI
AI Video Editor
Upload a short clip, describe the visual change, and turn the existing shot into a focused video edit.
MP4 / MOV · 2–10 sec50 credits / started secUpdated September 2026
IMG AI studio
Edit a clip with one clear instruction
50 credits / sec
Concept frame
Product color direction
One clip · 2–10 seconds · 20 MB max
1Describe the edit
One focused change is easier to review.137/2000
I own or have permission to edit this footage and will review the AI result before publishing it.
Your balance0 credits
This edit— credits
720p output · source duration mode · provider audio setting
The direct answer
What is an AI Video Editor?
An AI Video Editor changes existing footage from a natural-language instruction. You provide the video and say what should be different; the model creates a video-to-video result from the source instead of inventing a scene. It fits product variations and controlled visual directions.
IMG AI keeps the decision visible: upload a 2–10 second MP4 or MOV, check duration and aspect ratio, write one instruction, see the credit estimate, then compare source and result. The server validates the file, the task has a stable identity, and completed work remains in My Creations.
Re-rendering can shift fine text, faces, reflections, or fast motion. Name what should change and what must stay unchanged, then review before publishing.
Jobs this AI Video Editor is built for
Try a new scene treatment
Describe a new atmosphere or background while protecting the subject, framing, timing, and motion.
Prepare a social cut concept
Use an existing short shot as the anchor for one idea. Treat the result as a concept, not a finished edit or grade.
How to edit a video with AI
Upload a supported clip
Upload one MP4 or MOV up to 20 MB and 2–10 seconds. IMG AI checks container metadata, dimensions, and aspect ratio before reserving credits.
Describe one controlled change
State the change, then list what to preserve: subject, movement, timing, framing, lighting, or audio. The editor uses a 720p contract and keeps source duration.
Review and download
Compare source and edited states. Check motion, edges, text, reflections, and audio before downloading. Refreshing can recover the same task instead of creating a duplicate request.
Owned concept examples
What to inspect in a video edit
These storyboard concepts show three controlled edit directions. They are illustrative case assets for this page, not measured customer results or generated videos from a live task.
Storyboard conceptProduct color direction
Inspect glass highlights, cap edges, label placement, and the point where the bottle moves through the shot.
Prompt direction: Change the amber glass bottle to cobalt blue glass. Keep its shape, cap, background, camera movement, timing, and label layout unchanged.
Storyboard conceptBackground treatment
Review the subject boundary, pedestal contact, shadows, and smoke crossing the bottle before campaign use.
Prompt direction: Replace the dark smoky background with a pale blue studio background. Keep the bottle, cap, pedestal, camera movement, and timing unchanged.
Storyboard conceptAtmosphere variation
Compare exposure, material color, and mood. Treat the storyboard as direction until final footage is reviewed.
Prompt direction: Shift the studio atmosphere toward a restrained blue-hour look. Keep the device, hand position, camera path, framing, and duration consistent.
What this AI Video Editor actually does
The page exposes a focused source-clip editing workflow. The controls below describe the current contract rather than promising capabilities the provider or account does not expose.
Prompt-led video-to-video editing
The source is the anchor. The contract sends its URL, aspect ratio, duration, 720p output, audio, and watermark settings to the provider.
Cost and input facts first
The workbench shows duration, aspect ratio, file rules, and 50 credits per started second before the paid edit is submitted.
A recoverable account task
Every request has one identity and credit reservation. My Creations keeps the task, while download checks account ownership.
AI Video Editor vs timeline editor vs text-to-video generator
Choose by starting material. IMG AI fits a short source clip and controlled direction; a timeline editor wins for deterministic layers; text-to-video wins when there is no shot to preserve.
IMG AI AI Video Editor
A short existing clip needs a prompt-led visual edit.
Upload one clip, describe the change, see the cost, run one tracked task, compare source and result, then download or keep it in My Creations.
Input checks, price visibility, recovery, ownership, and the interactive flow are designed for a focused video-to-video change.
There is no timeline, keyframe, mask, or per-frame control, and fine detail can be re-rendered.
Open the IMG AI editorManual timeline editor
The change must be deterministic, layered, and hand-tunable.
Cut, mask, keyframe, grade, mix audio, and export on a timeline.
It offers control over timing, masks, typography, transitions, sound, and frame-level corrections.
A believable object or background transformation takes more setup and operator time.
Text-to-video generator
There is no source footage, or the shot can be invented.
Describe a scene, generate a clip, and iterate on composition, motion, and style.
It explores a shot from a blank page without source footage or an edit plan.
It is a poor starting point when the product, actor, camera path, or timing must remain recognizable.
Best inputs for an AI video edit
Start with a short clip where the subject is visible, the camera path is understandable, and the requested change can be stated in one sentence. Explicitly protect the subject, movement, timing, and framing. That gives the output a clear reference and gives you a fair checklist for review.
Limits you should know
- Input is one MP4 or MOV, up to 20 MB and 2–10 seconds. Dimensions must be 720–1,920 pixels and match 16:9, 9:16, 1:1, 4:3, or 3:4.
- The current editor is a focused 720p workflow, not a multitrack timeline, masking, rotoscope, or finishing suite.
- Video-to-video generation can shift text, logos, faces, reflections, hands, and fast edges. Review playback and important frames.
- The workflow keeps source duration and requests the configured audio mode; it cannot promise perfect preservation of every sound or subtitle.
Input, output, and pricing at a glance
The editor shows the same facts before submission, so you can decide whether this workflow fits the clip before spending credits.
- Input
- One MP4 or MOV
- Duration
- 2–10 seconds
- Maximum file
- 20 MB
- Dimensions
- 720–1,920 px per side
- Aspect ratios
- 16:9 · 9:16 · 1:1 · 4:3 · 3:4
- Output mode
- 720p video-to-video edit
- Credit rate
- 50 credits per started second
- History
- My Creations and owned download
Use footage you are allowed to edit
Upload only footage you own or are authorized to process. Check people, products, logos, music, subtitles, and commercial claims before publishing an AI-edited clip. IMG AI keeps the task tied to your account and protects the result download with ownership checks.
Choose the next workflow
AI Video Editor FAQ
What does an AI Video Editor do?
It edits existing footage from a natural-language instruction. Upload a short clip, describe the visual change, and receive a video-to-video result.
Can I edit a video online without installing software?
Yes. IMG AI handles upload, instruction, task tracking, review, and download in the browser. Sign in and use a 2–10 second MP4 or MOV up to 20 MB.
How should I write an AI video editing prompt?
Name the change, then say what remains unchanged. Example: “Change the bottle to cobalt blue glass. Keep its shape, cap, camera movement, background, timing, and label layout unchanged.”
How many credits does an AI video edit use?
IMG AI charges 50 credits per started second. A 2.1-second clip uses 150 credits because duration rounds up; the workbench shows the estimate before submission.
What video formats and dimensions are supported?
Upload one MP4 or MOV up to 20 MB and 2–10 seconds. Dimensions must be 720–1,920 pixels with 16:9, 9:16, 1:1, 4:3, or 3:4. The server reads metadata.
Will the original product, person, or text stay exactly the same?
No AI editor guarantees pixel-perfect preservation after re-rendering. The source anchors the result; inspect faces, labels, text, edges, motion, and audio before publishing.