AI tagging built for editors, not IT departments.
DAAAM doesn't just detect objects. It writes editorial descriptions: shot type, composition, mood, lighting, space for graphics, likely edit uses. Every frame. Automatically.
Every frame described and indexed on-device — 23,705 sampled moments across 5,365 assets in this library, no cloud round-trip.
What every frame gets
Most AI tagging tools return flat object labels: "person, tree, building." Useless for editorial work. DAAAM returns a structured editorial read — each field designed for someone who makes cuts.
| Field | What it captures | Example |
|---|---|---|
| scene | Dominant visual hook | "Two people at a café table; one leans forward mid-sentence" |
| shot_type | Editorial vocabulary | wide / medium / close / macro / aerial / POV / insert |
| composition | Spatial structure | "Subject rule-of-thirds left, significant negative space right" |
| space_for_titles | Zone map of usable areas | "Upper-right: clean sky gradient, ideal for lower thirds" |
| lighting_notes | Cinematographer's vocabulary | "Hard key from camera-left, warm orange cast, 4:1 contrast ratio" |
| movement_or_energy | How it feels in motion | "Handheld, slight drift, intimate and present" |
| mood_and_editorial_intent | Why an editor would pick this | "Tension before a reveal. Use before a dialogue cut" |
| likely_edit_uses | Timeline function | establishing / j-cut / b-roll / insert / reaction / cutaway |
| readable_text_in_frame | Legible on-screen text | "Street sign: Bourke St" |
| confidence_note | Uncertainty flags | "motion blur on subject; face partially occluded by hair" |
Wrong? Fix it in a click.
Every field is model-generated — and every field is yours to change. Click any caption or tag to rewrite it in your own words, add your own, or hit Re-analyse to regenerate. Your edits are never overwritten, so the catalogue ends up reading exactly the way you think about your footage.
Editorial descriptions, not object labels
Most AI tagging tools stop at what's visible: person, tree, building. DAAAM writes the way an editor would log a shot — composition, light, mood, motion, and where titles could sit. Every distinct moment in your footage, automatically.
Efficient on real-world footage
Interviews, locked-off wides, and hovering drone shots don't need to be re-read frame by frame. DAAAM spends its time on what actually changes — so a full day's rushes indexes in hours, not days, without skipping the shots you'll actually cut to.
Why this beats generic object detection
<strong>Generic AI tagging:</strong> person, chair, table, cup, window
<strong>DAAAM editorial tagging:</strong> shot_type: medium · composition: subject rule-of-thirds left, negative space right · lighting: warm orange key, blue window fill · space_for_titles: upper-right clean gradient · mood: tension before reveal · likely_edit_uses: reaction beat before dialogue cut
One tells you what's in the frame. The other tells you why you'd use it.