EdgeTAM rotoscope
pixel-accurate masking without keyframes
Create pixel-accurate rotoscopes automatically with EdgeTAM. EditAssist segments subjects at the pixel level, generates editable masks, and exports them to DaVinci Resolve, Premiere Pro, After Effects, and Final Cut. No green screen required.
Workflow
Click a subject in any frame and EdgeTAM builds a pixel-accurate segmentation that propagates through the clip. It handles hair, motion blur, transparency, and overlapping colours. Refine the mask with brush strokes or exclusion zones, then export as an alpha matte, a power window, or a keyed timeline.
The segmentation is local and frame-accurate. Use it for background replacement, selective grading, VFX compositing, or stylised effects. Export to any NLE as a matte layer, a pre-keyed clip, or animated shape data.
Prompts
“Rotoscope the dancer and replace the background with a city skyline.”
“Create a pixel-accurate mask of the spokesperson for a background replacement.”
“Segment the product from the table and export an alpha matte for After Effects.”
“Mask out the rigging wire from the stunt shot frame by frame.”
“Create a hair-accurate matte for the interview subject so I can grade the background separately.”
“Rotoscope the dog and apply a slow-motion blur only to the background.”
“Segment every person in the frame and create individual masks for each.”
“Build a clean plate by masking the moving subject and filling the background with adjacent frames.”
Features
Pixel-accurate segmentation
EdgeTAM segments subjects at the pixel level, handling hair, motion blur, and fine detail without chroma keying.
Propagation & refinement
Masks propagate across frames automatically. Refine with brush strokes, exclusion zones, or edge feathering.
Multi-NLE export
Export as alpha mattes, power windows, keyed clips, or shape data for Resolve, Premiere, After Effects, and Final Cut.
No green screen needed
Works on any background. No studio setup required. Local processing, no upload.
Local-first
All transcription, vision analysis, semantic search, and clustering run on your own GPU. Nothing about your footage is uploaded. The only data that leaves is the text of your conversation with the agent. It's free to download, local models are free, and a new account starts with £15 of credit and no card. For long-running cloud work, pay as you go, or choose £20/mo Starter or £100/mo Pro.
Getting started
- 01Download EditAssist for macOS or Windows and create an account. You get £15 credit and no card is needed.
- 02Ingest your footage. The agent transcribes, indexes, and clusters automatically.
- 03Describe what you want in plain English and let the agent build the cut.
- 04Review, refine with conversational commands, and export to your delivery spec.
More tools: AI object tracking, AI color grading, AI B-roll finder. Or see everything EditAssist connects to.