How does it work?
We transform your photos into animated mini-videos (≈ 6 seconds) using artificial intelligence, to bring your memories back to life — yours, your parents', and even your grandparents'.
Send your photo
When placing your order, upload your photos.
Accepted formats: JPG, JPEG, PNG.
Important recommendation (quality)
For optimal results: choose a borderless photo with clearly visible front-facing faces (avoid backs and profiles). The photo can be slightly blurry. See examples on the website.
✅ Good to know
- • Old photos (even slightly yellowed ones) work very well.
Our AI animates your photo
Our AI analyzes the photo (face, gaze, expressions, lighting) and generates subtle, natural movement.
The goal: a "living" result without odd effects.
⚠️ Important things to know
- • Slightly blurry photos accepted: we can often achieve a good result, but sharpness helps a lot.
- • Front-facing face recommended: if the face is in profile, hidden, or turned away, the AI may need to "reconstruct" details, which can alter the likeness.
- • The more obstacles there are (highly reflective glasses, hands in front of the face, low light), the more the result may vary.
Receive your mini-video
Once ready, you receive your MP4 mini-video (≈ 6 seconds) by email, via a secure link.
It is ready to:
- • be shared (WhatsApp / social media)
- • be displayed on a connected frame
- • be kept as a digital keepsake
🔒 Privacy
Your files remain private. You certify that you have the rights/consents for the photos you send.
How does artificial intelligence work?
Our system uses models specialized in image animation.
- → Image analysis: detection of key areas (faces, contours, expressions)
- → Movement generation: creation of coherent movement from visible details
- → Limitations: the AI can only properly animate what it "sees". Missing information (profile/back view) may be interpreted.
Tips for a top result
Scanned photo
More detail, cleaner result
Clearly visible face
Front-facing if possible
Good resolution
Sharper = more faithful
Avoid profiles / hidden faces
Risk of less accurate likeness
Even lighting
Avoid harsh shadows on the face
(≈ 6 seconds — MP4 — secure link)
What exactly does our artificial intelligence do?
Our photo artificial intelligence starts by analyzing the image you send us. It detects faces, their orientation, eyes, mouth, facial contours, and certain key expression points. This step helps understand what is already present in the photo, without altering the person's identity. The goal is not to recreate someone, but to give a slight movement to an existing image.
Next, the AI generates a very short, subtle face animation. The result may give the impression of a gaze gently shifting, a smile forming, or a slight head movement. People often talk about a moving photo, but it is important to understand that this is not a complete scene. The AI does not turn a photo into a film. It creates a keepsake mini-video of a few seconds, designed to remain natural and respectful.
Our approach is based on ethical deepfake technology. This means the AI is not used to deceive, impersonate an identity, or make someone say something. It does not change the face, invent a new scene, or add absurd or sensitive elements. No invented background, no political message. The idea is simple: bring a family photo to life, not distort it.
Tips for getting the best result
Front-facing photo
To understand how to animate a photo, imagine that the AI needs to "read" the face. A front-facing photo generally works better than a strong profile. If the person is looking at the camera, the eyes, mouth, and contours are easier to detect. The resulting movement then appears more natural.
Good lighting
A well-lit photo often gives a cleaner result. If the face is too dark or backlit, the AI will have more difficulty detecting it correctly. Soft light, such as near a window or outdoors without direct sunlight, helps achieve a smoother and more faithful animation.
Sufficient resolution
The sharper the photo, the more likely the final result is to be convincing. An image that is too small, blurry, or heavily compressed limits the possibilities of artificial intelligence. Ideally, send the original photo when you have it. Avoid poor-quality screenshots or images recovered after several shares.
Clearly visible faces
Faces should be as unobstructed as possible. Sunglasses, a hand in front of the mouth, a cropped face, or a person placed too far away can reduce the quality of the result. For face AI animation, the AI must be able to correctly identify the eyes, nose, mouth, and overall shape of the face.
How long does the process take?
The usual timeframe is 24 to 48 hours. This includes photo processing, generation of the keepsake mini-video, and then a human review before delivery. This review is important: it ensures the result remains natural, respectful, and true to the spirit of the original photo. We prefer to take a little time rather than automatically sending an animation that would not be satisfactory.
What do you actually receive?
You receive a video file in MP4 format, usually about 6 seconds long, delivered via a secure link. This file can be kept, shared with loved ones in a personal setting, or displayed on a compatible WiFi-connected digital photo frame, such as a Frameo frame with a 10-inch touchscreen.
Frequently asked questions
Are my photos stored after delivery?
Your photos are used only to create the requested animation. They are not used to publicly train an AI or to create other content without your consent.
Can I request a revision if I am not satisfied with the result?
Yes, if the result has a visible issue or does not respect the spirit of the photo, you can request a revision. We then check what can be improved.
Which image formats are accepted?
The most common formats are accepted, including JPG, JPEG, and PNG. For a better result, choose a sharp, non-blurry photo sent in the best possible quality.
Is the result always successful?
No, not always. The AI depends heavily on photo quality. A blurry, too dark, or partially hidden face can produce a less natural result. In that case, another photo may be preferable.