OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Image & Video AI

4Dface

4Dface is an open-source C++ demo app for tracking faces and fitting a 3D Morphable Model to live webcam or video input.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

4Dface — OpenViST project profile in Image & Video AI.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use the repository when you want to build and explore the documented research demo for video-based face tracking and 3D model fitting. For a more current commercialised version, the README directs readers to 4dface.io.

Who it is for

  • Developers building or studying computer-vision applications that process face video.
  • Researchers exploring 3D face-model fitting and landmark-based tracking.

Documented uses

  • Track faces in webcam or video footage.
  • Fit a 3D Morphable Model to facial input.
  • Process a video file and save the current fitted model as an OBJ file.
Before you choose

Requirements & limits

The repository is public and licensed under Apache-2.0. Its README states that the included 3D morphable face model is free for non-commercial use; commercial use requires contacting the Centre for Vision, Speech and Signal Processing.

  • The repository version is a 2015/2016 research demo, with the README describing real-time performance as around five frames per second.
  • The documented stack includes Boost, OpenCV, Eigen, CMake and a recent compiler.
  • The README identifies eos and superviseddescent as underlying libraries and recommends building them separately for library-level work.
  • The included morphable face model has a stated commercial-use restriction distinct from the repository's Apache-2.0 license.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Build and run the demo on a video

Prepare and build the repository, then run the installed app on a video file.

Before you start

  • A recent compiler: GCC 5 or newer, Clang 4 or newer, or Visual Studio 2017 or newer.
  • Boost 1.54 or newer, OpenCV 3.0 or newer, Eigen 3.3.0 or newer, and CMake.
  • The repository cloned with submodules, or its submodules initialized afterward.

Steps

  1. Clone the repository recursively, or run the command below inside the cloned directory.

    Documented command
    git submodule update --init --recursive

  2. Install Boost, OpenCV and Eigen; on Windows, the README recommends vcpkg.

  3. From outside the source directory, create a build directory and configure it with CMake. On Windows, use the documented Visual Studio generator and configure initial_cache.cmake or a vcpkg toolchain as needed.

  4. Build with make or Visual Studio, then install with the command below or the Visual Studio INSTALL target.

    Documented command
    make install

  5. Run 4dface -i videofile from the install directory to process a video file.

Expected outcome

The installed demo processes a supplied video file for face tracking and 3D Morphable Model fitting.

Guide limitations

  • The repository describes itself as a 2015/2016 research demo and estimates its real-time performance at around five frames per second.
  • The included 3D morphable face model is free for non-commercial use; commercial use requires contacting the Centre for Vision, Speech and Signal Processing.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
17 October 2026
Original website title
patrikhuber/4dface · Official homepage title · retained in its source language
How we review sourcesSuggest a correction