DeepMotion
Browser-based AI motion capture: video or text in, 3D animation out.
About
DeepMotion's Animate 3D converts uploaded video into rigged 3D character animation through the browser — AI pose estimation plus physics simulation (foot locking, weight, hand tracking) with no suits, cameras, or software installs. Output goes to FBX, BVH, GLB, or MP4 for Unity, Unreal, Blender, and Maya, with a Rotoscope Pose Editor for fixing tracking mistakes frame by frame.
SayMotion, its text-to-animation beta, generates motion from prompts without any video input. Pricing is freemium (free tier capped at ~60 seconds/month; paid from ~$17–19/mo). Honest limits: tracking degrades with occlusions, loose clothing, and complex multi-person scenes, and the full API/SDK path is sales-gated rather than self-serve.
What it can do
- Video-to-3D-animation via browser (Animate 3D)
- Text-to-animation beta (SayMotion)
- Physics simulation, foot locking, hand and face tracking
- FBX, BVH, GLB, MP4 export
Setup
Sign up at deepmotion.com; upload a video in the browser-based Animate 3D dashboard.
Prerequisites
A clear full-body video (2–6m camera distance, neutral lighting); paid plan for longer clips.
Use cases
Works with
Trust score — 7/10
Scored 7/10 (metrics tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 63; Trending score 70/100 from researcher assessment.
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.