Find the best moments. Send them to your NLE.
AI video search for post-production teams. Imaginario indexes your rushes, dailies and archive where they live and understands every shot: what is said, who is on screen, what happens.
Search it all, pick the moments, and send clips and sequences straight into Premiere Pro, DaVinci Resolve or Avid. The originals never move.
Trusted by hundreds of video marketing and editing teams
The content post-production teams run through Imaginario
Dailies, the archive, the b-roll, the string-outs and the interview-led cut. What gets indexed, and what the edit does with it.
| Type of content | Use case |
|---|---|
| 01Dailies and rushes from productionCamera A, camera B, drone, GoPro and the audio recorder |
|
| 02Finished programs and the archiveYears of programs and unused rushes as one library |
|
| 03B-roll from past projectsIt usually already exists somewhere in an earlier one |
|
| 04Selects, string-outs and rough cutsPulls, selects reels, assemblies, compilations, montages and thematic reels |
|
| 05Interview-led documentaries and realityPaper edits, radio edits and story spines from the soundbites |
|
Your content stays private and is never used to train generative models. See all the ways to search and the workflows post-production teams run.
In the trade press
“Imaginario AI vastly simplifies and enhances the content creation process, empowering post-production and marketing teams of all sizes to distribute more content to a wider audience and better monetise their assets.”
Built for the rushes, not just the archive
Search every rush and daily for anything
Let our AI watch your videos
Point Imaginario at the project folders on your shared storage, whether that is Avid NEXIS, a NAS or Amazon S3. It works from 3 to 5 Mbps proxies at any resolution, so ProRes and camera originals never move, and it logs speech, visuals, people, actions and locations on every clip at shot level. Your footage stays private and is never used to train generative models.
Understand the whole conversation
Timecoded transcripts with speaker identification, captions and chapters in more than 100 languages are the start. Search interviews by theme, topic or emotion with dialogue search, pull every take where the contributor mentions the accident, and hand the producer the soundbites in order for a story spine.
Finally, usable visual and audio search
Visual search, audio search and person search run at one frame per second, so every wide shot of the farmhouse at dusk, every named contributor and every sound is findable across the whole project.
Locate specific moments or perfect B-roll in seconds.
Find every person and every face
Every contributor, presenter and crew member who appears on camera is identified across the whole project, so one search returns every take and every shot they are in, on any camera and any day of the shoot. Person search works from a face bank you build in minutes, recognizes public figures on its own, and reads facial expressions and emotions, so the reaction shot you need is a search away.
Select, assemble, send to the edit
Highlight a passage in the transcript and Highlight to Edit turns it into a clip on the spot. Save takes and search results into collections, let multi-video stories in StoryLab assemble a first pass, then export a clip or a whole collection as a sequence with timecoded metadata into Premiere Pro, DaVinci Resolve or Avid Media Composer.
With a paid subscription, you can export an unlimited number of HD clips every month.
Beyond tagging: human-level understanding of every shot
Tags tell you a shot has a car in it. Imaginario understands the shot the way an assistant editor would: who is in it, what they say, what happens, where, in what mood, and how it fits the story. Every rush gets this treatment automatically, in more than 100 languages. This is what the search and the metadata you send to the edit are built on.
For Enterprise customers, we can create custom models based on your footage and map outputs to your own vocabulary or editorial taxonomy.
Plugs into the suite you already run
Every leading NLE, shared storage system and MAM. On-prem storage connects through Cubix from our partner Ortana, cloud storage is self-serve, and every storage connection works in both directions, so finished clips go back where the team expects them.
Frequently asked questions
Do our ProRes originals have to move?
No. Imaginario works from proxies at 3 to 5 Mbps in any resolution, and indexes the folders you choose on your shared storage. The camera originals, the folder structure and the permissions stay exactly where they are.
How does Imaginario connect to our storage?
Most on-prem connections run through Cubix, the media orchestration platform from our partner Ortana, set up once with your storage admin. Cloud storage such as Amazon S3 is self-serve: add the bucket and the access keys your cloud team creates, and add as many buckets as you need. Every storage connection works in both directions.
What arrives in Premiere Pro, DaVinci Resolve or Avid?
Clips, sequences and timecoded metadata, ready to edit. Export a clip from a single asset, or a collection across many assets as a sequence for a compilation or rough cut. The steps are the same for Premiere Pro, DaVinci Resolve and Avid Media Composer.
Can it build a rough cut or a story spine?
Yes. Select the soundbites and takes that carry the story, in order, and multi-video stories in StoryLab assembles a first pass you export as a sequence. Editors finish the cut in the NLE. Ask your library answers questions across the project, such as which takes have the cleanest audio.
Which productions get the most out of it?
High shooting ratio genres: factual documentaries, factual entertainment, reality and interview-led series, where a handful of episodes come out of hundreds of hours of rushes. The same index covers the archive, so B-roll and moments from earlier projects come back into a new edit without anyone remembering where they were.
We run a MAM. Does Imaginario replace it?
No. Imaginario sits next to your MAM, it does not replace it. It indexes the library the MAM manages and returns time-based and asset-level metadata to it, mapped to your fields with taxonomy and schema mapping, so search inside the MAM gets better too. See the MAM and DAM integrations.
Get started today
Enjoy thousands of minutes of AI analysis, transcriptions, search, and auto-captions, plus much more.