Group recurring people
Automatically collect appearances of the same person across interviews, multicam footage, episodes, and archives.
Jumper Face Recognition
Find every appearance of a person in your footage
Jumper detects and groups recurring faces across analyzed media, helping editors locate cast members, interview subjects, speakers, and other people without manual logging.
What it does
Face Recognition in Jumper
Face recognition in Jumper is designed for people in an editor’s own footage, including non-celebrities. Jumper detects faces, groups likely matches, lets the editor assign names, and makes those people available as search filters. Processing stays on the local device.
Core capabilities
Built for real editing workflows
Name and search people
Assign a name to a detected face group, then use that person as a filter when searching footage.
Protect identity data
Detection, grouping, and search happen locally rather than sending private footage to a cloud service.
How it works
From source footage to useful result
- 1
Analyze faces
Run face analysis on the media relevant to the project.
- 2
Review and name groups
Confirm useful groups and assign names to the people the edit needs to track.
- 3
Search by person
Find that person’s appearances directly or combine the person filter with a visual search.
Technical documentation
Frequently asked questions
Can Jumper recognize people who are not celebrities?
Yes. Jumper groups similar faces found in the editor’s own footage. The editor can then name useful groups and search for those people.
Is face recognition processed in the cloud?
No. Face detection, grouping, and search run locally on the editor’s device.
Which Jumper plan includes face recognition?
Face recognition and people search are available with Jumper Pro.