Video analysis for research

See how learning happens.

MASS turns hours of observational video into structured, shareable insight. An AI agent guides your analysis, a collaborative labeler captures what's happening on screen, and the participation map makes patterns visible at a glance.

Built for education researchers. Ready for any video-based activity analysis.

PARTICIPATION MAP — Classroom Session 04
Student A
Student B
Student C
Group work
00:0010:0020:0030:00
LABELER — session_04.mp4 · pass: group-work
Student A · on-task
Student B · discussing
ATLAS
Find every segment in this week's recordings where students are working in groups.
I analyzed 12 videos in Dataset: Week 14 and found 47 candidate segments of group work. I'd like to create an annotation pass called "group-work" so you can review them in the labeler.
Create annotation pass "group-work"?
How it works

From raw footage to research insight

Three steps, with an AI agent helping at every one of them.

01

Upload & organize

Bring in classroom recordings, lab sessions, or field footage. Organize them into projects and datasets, or just tell the agent what you have and let it set things up.

02

Annotate with AI assistance

Ask the agent to scan videos for the activities you care about, then refine its suggestions in the labeler: bounding boxes, subject tracking, and click-to-segment masks.

03

Visualize & share

The participation map lays every coded moment on an interactive timeline. Filter, group, and compare, then share a live view with collaborators or stakeholders via a link.

Built for education research

Understand what's really happening in the classroom

Observational video is the richest data education researchers have, and the most time-consuming to analyze. MASS was designed to close that gap: code student behavior systematically, keep your team consistent, and surface the patterns that field notes miss.

  • See how participation unfolds across a session: engagement, collaboration, discussion
  • Code group work, teacher-student interaction, and transitions between activities
  • Built-in review workflow mirrors second-coder practice: submit, request changes, approve
  • AI-suggested annotations accelerate coding without removing the researcher from the loop
  • Compare across sessions, classrooms, and conditions on a single visual timeline
  • Share findings as interactive views, not static screenshots in a slide deck
Not just classrooms

Anywhere activity happens on video

The same workflow — annotate, track, visualize — applies to any research question you can point a camera at: usability studies, behavioral observation, sports analysis, workplace studies, and more.

See the use cases
Real-time collaborationMultiple coders on one video, conflict-free, with per-subject editing locks.
Review & approval built inEvery annotation set moves through a clear submit → review → approve cycle.
You approve every AI actionThe agent proposes; the researcher decides. Sensitive actions are gated and audited.

Ready to see your data differently?

We're working with early research partners now. Tell us about your study and we'll show you MASS on your kind of footage.

Get in touch