Feature · Video

Find the video, not just the photo

Your videos are the part of your library you can never search. You remember filming the birthday, the cycling race, your kid's first steps, but the clips sit buried in folders named after dates, unwatched. PhotoCore now indexes your short videos right next to your photos, so you can describe a moment and land on the exact second it happens, all on your Mac.

Describe a clip, find itPlays at the exact secondFaces inside videosFully offline
The problem

Videos are the black hole of your library

Photos you can at least skim as thumbnails. A three-minute clip you have to open and scrub, so you never do. The result is that the most alive moments you own, the ones with movement and sound, are the hardest to ever find again. Search stops at the edge of your videos.

PhotoCore closes that gap. Your short clips are indexed with the same local AI as your photos, so ‘describe it and find it’ finally works for video too.

How it works

Describe it. Land on the moment.

The clip, and the exact second

Turn a folder into search and PhotoCore samples each video, understands every moment, and matches it to your words. It returns the clip and opens playback right at the second that matched, so you skip the scrubbing entirely. It is the same natural-language search as the rest of the app, in more than 100 languages.

  • Type what happens, not a filename
  • Playback opens at the matching second
  • Short clips indexed next to your photos
PhotoCore — video results
Faces in video

Know who is in the clip

Nobody else does this offline

PhotoCore looks through a video for faces, picks the clearest frames, and recognizes who is in it. The one identity you named across your photos now spans your videos too, all on the Neural Engine, nothing uploaded.

  • Finds and reads faces frame by frame
  • Uses the same person you named in photos
  • Runs on device, no cloud

@mentions and people filters, for clips too

Type ‘@Anna’ and the clips she appears in show up beside the photos. Filter by how many people are in frame, or stack a name with a description. Video becomes a first-class citizen of search instead of a folder you dread opening.

  • Mention a person inside any search
  • Filter clips by people count
  • Pairs with the quality and date filters
Built to last

Fast library, no babysitting

Short clips, in focus

PhotoCore concentrates on the clips worth finding and skips anything too long or too large, so indexing stays quick and the library never bogs down.

Nothing stalls

A per-file watchdog and a resume journal mean a single slow file or a sleepy NAS can never block the rest of your indexing.

Reindex brings them in

Add a mixed folder or reindex one you already have. The clips PhotoCore skipped before come straight in, your photos untouched.

Offline & private

Works with the drive unplugged

Findable offline, like your photos

Video thumbnails, search results and playback previews keep working when the external drive or NAS is disconnected, exactly like the rest of your archive. Index once, keep finding forever.

  • Thumbnails and results survive a disconnect
  • Same behaviour as photos on external drives
  • Reconnect to play the full clip

On device, nothing uploaded

Every frame is analysed locally on the Apple Neural Engine. There is no cloud, no account and no tracker, so your home videos and the faces in them never leave the Mac.

  • Local AI on Apple Silicon
  • No upload, no account
  • Face data deletable in one click
FAQ

Frequently asked questions

Can I really search videos by what happens in them?
Yes. PhotoCore samples each short video, understands every moment with local AI, and matches your words to it. Type ‘birthday at the table’ or ‘cycling race in the mountains’ and it returns the clip, then opens playback at the second that matched, no scrubbing.
Does it recognize faces inside videos?
Yes, and it does it offline. PhotoCore looks through a clip for faces, picks the clearest frames, and recognizes who is in it. Name a person once and they turn up in your videos too, with @mentions and people filters. Nothing is ever uploaded.
Will indexing videos slow my library down?
No. Videos are indexed with smart limits: PhotoCore focuses on short clips and skips anything too long or too large, a per-file watchdog stops a slow file from stalling the rest, and a resume journal picks up where it left off.
Do my videos or faces leave the Mac?
Never. All analysis runs locally on the Apple Neural Engine. There is no cloud, no account and no upload, so your videos and any face data stay on your machine.
How do I add videos to a library I already indexed?
Just reindex the folder, or open one that mixes photos and videos. PhotoCore detects the clips it skipped before and brings them in, without re-processing your existing photos.
Get started free

Bring your videos into the search

Download PhotoCore free and index up to 200 photos and clips to try video search and faces in video. Full access is a one-time purchase, with no subscription, ever.