How many photos do you need to create a living memory?

The short answer is simple: one clear photograph can be enough to begin. But if you have several useful photographs, three to five varied images often give us a better understanding of the person or pet.

Useful beats numerous

Ten photographs taken seconds apart do not necessarily teach us more than two good ones. What helps is variety.

• A clear front-facing image can establish facial structure.
• A side angle can show profile and head shape.
• A candid image may reveal posture or the natural smile.
• A photograph from the correct time period can help with age, hair and clothing.

What if the pictures are old, blurry or imperfect?

Send the best material you genuinely have. Old family photographs were not created for an AI workflow, and that is completely normal. A slightly imperfect photograph can still contain useful identity information.

What matters is that you tell us when something is uncertain. If a photograph is from the wrong age, if the hair had changed, or if a particular picture does not show the person’s usual expression, add that context.

Can I mix photographs from different years?

Yes. The age of the source photographs and the age of the memory do not need to match perfectly. A photograph closer to the intended period usually reduces the amount of estimation required, but mixed-age references can still be useful when the desired age is clearly specified.

This is particularly common with pets and older family photographs, where the best available images may span many years.

What about video and voice?

Short videos can provide useful movement reference. A separate clear voice reference can be useful if spoken dialogue is being considered, but voice is a separate production decision and requires appropriate permission. Your storytelling recording is not automatically treated as a voice sample.