About Oceanir
Physical intelligence for the world
Find a scene. Find where it was taken.
Oceanir works out what is true about the physical world from what a camera recorded. We start with place, because knowing where something is connects it to everything around it.
What we make today
Oceanir app
Find a scene. Find where it was taken.
Add a photo, a frame from a video, or a link to a post. You get the likely places, the clues behind each one, and what is still uncertain. Free to start, no AI know-how needed.
Manta, for organizations
A visual evidence model for work that gets questioned afterwards.
Manta establishes where a photograph was taken from its content alone, resolves it toward the street or the building, and returns the evidence that led there. Access requires organization approval.
How we answer
- Evidence comes with every answer
- You see the clues for a place and the ones against it, and the alternatives that were ruled out. An answer you can't check is only a guess.
- We say when we're unsure
- A confident result and a weak one look different. When the evidence doesn't support a street, we don't name one.
Where it goes
Place is the first fact we can prove. The same evidence leads to the next questions about a scene, across photos, video and live sensor feeds, and toward robots and vehicles that need to locate themselves.
Where
Where a photo or a video was taken. This is the part we ship today.
When
When it was taken, read from the scene itself.
What changed
What is there now and what has changed since.
Who uses it
Anyone who sees a place and wants to know where it is. And the people whose answer gets questioned later: investigators, journalists and editors, legal teams, insurance claims and trust and safety teams.
Have a photo? See where it was taken.