The same character, every shot
A portrait or character sheet anchors the face, hair, outfit and build, so a person can walk into a new location or a new story beat without turning into someone else.
Keep what matters · Change everything else
Reference to Video AI Generator
With reference to video, you give the model images, clips or audio for the things that must stay the same — a face, a jacket, a product, a camera move — then describe a brand-new scene. The references hold the identity; the prompt writes the shot.
- Same face in every shot
- Mix images, video and audio
- Product details preserved
You need an account to generate. Each run spends credits according to model, duration and resolution, and the exact cost is shown on the button. Failed tasks are refunded automatically.
Reference to Video Examples: Board, Then Scene
Each example pairs the reference board that was uploaded with the video it produced. Look for what carried over — faces, costumes, products — and what the prompt changed.
Reference board- Create a single vertical anime animation development board for an original emotional short film titled "The Girl and the Lighthouse". The output must be ONE combined image containing an anime character design sheet and a cinematic storyboard page. IMPORTANT: → Create fully original anime characters → Do not imitate any existing anime studio style directly → Do not resemble any copyrighted anime characters → Keep designs unique, soft, nostalgic, and emotionally warm STYLE: → Premium anime pre-production board → Mix loose hand-drawn storyboard sketches with semi-rendered anime keyframes → Use soft watercolor-style lighting, cool sea blue and warm lighthouse gold colors, gentle shadows, red storyboard borders, blue motion arrows, handwritten production notes, timing notes, and lens notes LAYOUT: Clean vertical composition with 2 sections. SECTION A — CHARACTER DESIGN SHEET Girl: → A gentle anime girl with long windswept dark hair, soft blue-grey eyes, and a quietly waiting expression → She wears: a soft cream knit sweater, a faded blue skirt, worn white canvas shoes, and a small green scarf → She should feel: patient, quietly hopeful, emotionally deep, nostalgically waiting → Show: front view, side profile, back view, 3/4 angle → Expressions: patient, soft smile, quietly hopeful, calm and waiting → Pose standing at the cliff edge watching the lighthouse → Pose sitting on cliff rocks with knees drawn up → Pose holding her scarf as the sea wind blows Lighthouse: → An old weathered lighthouse on a rocky seaside cliff — tall, worn stone, warm rotating golden light at the top → It should feel: steady, faithful, warm, a symbol of patient waiting and guidance → Show: wide view from cliff, close-up of rotating light, reflection on sea water Add tiny handwritten notes and soft pastel anime color swatches. SECTION B — STORYBOARD Create 8 cinematic anime storyboard panels in a neat grid. Each panel should include handwritten camera notes, blue motion arrows, timing notes, and lens notes. Story Beats: → Wide shot of an old anime seaside village at dusk — calm sea, rocky cliffs, warm lighthouse light beginning to glow → The girl walks alone along the cliff path toward the lighthouse as the sun sets → She sits down on the cliff rocks and looks out at the sea, quiet and waiting → Close-up of her face — patient, softly hopeful, emotionally still → The lighthouse light begins its slow rotation, sending warm golden light across the sea → She watches the light sweep across the water — each rotation steady and faithful → Wide shot — girl on the cliff, lighthouse above, the vast calm sea below, warm dusk light → Emotional ending — the lighthouse light sweeps across her face one more time, warm and gentle, the whole sea glowing behind her Environment: Old anime seaside village at dusk. Rocky cliffs, calm sea, old stone lighthouse. Warm golden dusk light on the water. Sea wind, distant waves, peaceful nostalgic coastal atmosphere.
New video
Lock the Look, Direct the Action
Image to video starts from one frame. Reference to video uses several inputs as a brief, so the same character or product can appear in a scene that never existed.
Reference to Video AI Models
Seedance 2.0 Mini is the default for quick 480p drafts. Seedance 2.5, Gemini Omni Flash, HappyHorse 1.1, PixVerse v6 and Vidu Q3 each accept references too, with different limits on how many files, which media types and how long the clip can be.
Why Use Our Reference to Video AI
The hard part of AI video is keeping things the same between shots. References are how you pin them down.
Several inputs, each with a job
Mix images, a source clip and audio in one request and say what each one is for — identity, wardrobe, product, camera motion or voice — instead of hoping the model guesses.
Products that survive the scene
Upload the real item from a couple of angles and its shape, materials, colours and label stay recognisable even in an ambitious lifestyle or cinematic setting.
New performance, no reshoot
Keep an approved subject and look, and change only what happens: the action, the location, the lighting or the camera path, all from the prompt.
How to Use the Reference to Video Generator
Add references, assign their roles, render. Being explicit in step two is what makes the difference.
Upload the references
Add the images, clips or audio that define the character, product, style or motion. How many files and which types you can add depends on the model.
Tell the model what each does
Write it plainly: "Image 1 is the woman's face, Image 2 is her coat, Video 1 sets the orbiting camera." Then describe the new scene.
Render and compare
Generate a short take and check the details you meant to keep. If something slipped, restate it in the prompt before rendering a longer version.
Reference to Video Use Cases
Anywhere continuity matters more than a one-off shot, references are the right tool.
Recurring characters
Keep an original character looking the same across a series of shots or episodes, without re-describing every facial and wardrobe detail in each prompt.
Product campaigns
Place one product in a studio, a kitchen and a cinematic street scene while its shape, finish and packaging stay as close to the real thing as possible.
Virtual creators
Produce short lifestyle clips for a digital persona. The identity references stay fixed while outfits, places and activities change from post to post.
Fashion and lookbooks
Reference the model, the garment and the styling separately, then direct walks, turns and close-ups to build a full lookbook sequence.
Storyboards into scenes
Use a storyboard panel for composition, a character sheet for continuity and the prompt for timing, camera and sound — a quick path from boards to moving shots.
Consistent brand worlds
Reuse approved colours, sets, materials and product photos across a family of videos, so every clip looks like it belongs to the same campaign.
More Reference to Video Examples
Browse prompts from the gallery and the clips they produced. Load one into the generator, add your own references and rewrite the scene around them.
Check Consistency Before You Scale
Every run shows its credit cost up front. Confirm that faces and products hold on a short draft first, then spend on the longer, higher-resolution take.
Reference to Video: Common Questions
What is reference to video?
It is a way of generating video where you supply reference files — images, video clips or audio — that define things the result must keep, such as a person's face, an outfit, a product or a camera move. The prompt then describes a new scene, and the model builds it around those references.
How is it different from image to video?
Image to video uses one picture as the literal first frame and animates it. Reference to video uses one or more files as guidance rather than as a starting frame, so the output can have a completely different composition, location and action while keeping the referenced details.
What can I upload as a reference?
Images are supported by every reference model. Some models also accept short video clips — useful for motion or camera paths — and audio for rhythm or voice. The upload area shows exactly which types and how many files the selected model accepts.
How do I keep a character consistent?
Use clear, well-lit references that show the face and outfit, ideally from more than one angle. In the prompt, name which reference is the character and restate the key details you care about. Keep the new action simple for the first take.
Can I use my own product photos?
Yes. Photos on a clean background from two or three angles work best. Tell the model which image is the product and ask for it to stay unchanged, then describe the setting and camera around it.
Can a reference video control the camera?
On models that accept video references, yes. Upload a clip with the movement you want and say in the prompt that it sets the camera path or the motion. The model follows its rhythm while rendering your new subject and scene.
Does reference to video generate sound?
Several of the reference models can generate native audio, and some accept an audio reference for voice or rhythm. If the selected model has no audio support, the audio switch is hidden and the clip comes back silent.
Why did the model ignore one of my references?
Usually because its role was not stated or it conflicts with another input. Refer to each file by its position — Image 1, Image 2, Video 1 — say what it controls, and remove references that are not essential to the shot.
Which model should I choose?
Start with Seedance 2.0 Mini for a cheap check. Seedance 2.5 is the most flexible with mixed media references; Vidu Q3 and PixVerse v6 are good alternatives for stylised or character-driven shots; Gemini Omni Flash also supports video references. Compare two on the same brief.
Is there a watermark?
Open Sora does not add its own watermark to generated videos.
Can I try reference to video for free?
You can open the tool and prepare references and prompts without paying. Generating uses credits; when welcome credits are enabled, new accounts receive a starter balance. The cost of each run is shown before you confirm.
Can I use the results commercially?
Commercial use follows your plan and the terms of service. You must have the rights to every reference you upload — including photos of people, products and artwork — and to anything the prompt asks the model to depict.
Same Subject, Any Scene
Upload the face, product or style you need to keep, give every file a job, and direct a new shot around it.
Generate From References