FrankMocap is a monocular 3D human capture system that estimates body, hand, and optionally face pose from a single RGB image or video. It regresses parametric human models (e.g., SMPL/SMPL-X) directly, producing temporally stable meshes and joint angles suitable for animation or analytics. The pipeline couples a robust 2D keypoint detector with 3D mesh regression networks and priors that keep results anatomically plausible. It can run frame-by-frame or with temporal smoothing, and includes demo apps for live webcam capture as well as batch processing. Outputs include textured meshes, joint locations, and model parameters that can be exported to common DCC tools and game engines. The codebase offers pretrained models, clear inference scripts, and utilities to visualize results, making single-camera motion capture approachable on commodity hardware. Researchers and creators use it for motion studies, AR/VR prototyping, character animation, and human-in-the-loop editing.

Features

  • Single-view 3D body and hand reconstruction using SMPL/SMPL-X
  • Image and video inference with optional temporal smoothing
  • Pretrained checkpoints and ready-to-run demo applications
  • Export of meshes, joints, and parameters for DCC and engine workflows
  • Utilities for visualization, rendering, and result inspection
  • Works on commodity hardware with straightforward setup

Project Samples

Project Activity

See All Activity >

License

MIT License

Follow FrankMocap

FrankMocap Web Site

Other Useful Business Software
Go From AI Idea to AI App Fast Icon
Go From AI Idea to AI App Fast

One platform to build, fine-tune, and deploy ML models. No MLOps team required.

Access Gemini 3 and 200+ models. Build chatbots, agents, or custom models with built-in monitoring and scaling.
Try Free
Rate This Project
Login To Rate This Project

User Reviews

Be the first to post a review of FrankMocap!

Additional Project Details

Programming Language

Python

Related Categories

Python 3D Modeling Software, Python Artificial Intelligence Software

Registered

2025-10-07