GV

GeezerVison

The World Wide Web, restored to 800 pixels of glory.

Home|About|Gaming cabinet|Archives|How it works|Starter sites

On this site

  • Sign in
  • ActionsCI/CD and automation
  • CodespacesCloud development environments
  • All Features
  • Documentation
  • Training
  • Certifications
  • Build with Copilot
  • Customize Copilot
  • Integrations
  • Trending repos
  • GitHub Sponsors
  • Developer articles
  • Enterprises

This view

  • Original page
  • New address

GeezerVison is showing github.com — GitHub - HRLTY/DenseFace3D: a software for the reconstruction of a real-time 3D face model from a monocular camera.

BranchesTags
Open more actions menu

Latest commit

 

History

25 Commits
25 Commits

Folders and files

NameName
Last commit message
Last commit date
picture
picture
 
 
.gitignore
.gitignore
 
 
CMakeLists.txt
CMakeLists.txt
 
 
DenseFace3D.cpp
DenseFace3D.cpp
 
 
OGL_OCV_common.cpp
OGL_OCV_common.cpp
 
 
OGL_OCV_common.h
OGL_OCV_common.h
 
 
README.md
README.md
 
 
ThinPlateSpline.h
ThinPlateSpline.h
 
 
conn.txt
conn.txt
 
 
glm.cpp
glm.cpp
 
 
glm.h
glm.h
 
 
landmark.txt
landmark.txt
 
 
linalg3d.h
linalg3d.h
 
 
ludecomposition.h
ludecomposition.h
 
 
readcsv.cpp
readcsv.cpp
 
 
samplePoint.cpp
samplePoint.cpp
 
 
samplePoint.h
samplePoint.h
 
 
stdafx.cpp
stdafx.cpp
 
 
stdafx.h
stdafx.h
 
 
targetver.h
targetver.h
 
 

Repository files navigation

DenseFace3D

This is a software for the reconstruction of a real-time 3D face model from a monocular camera by Rui Huang.

Demo: Youtube or Tencent Video

Feature:

  • Real-time landmark detection, mesh densifying (spline interpolation), triangulation and texture mapping for each frame.

  • Two stages buffering — Both the landmark detection and spline interpolation can fail and the last valid instance is kept in those cases.

  • Fixed location, stabilization and auto-scaling — The the capture is robust to face movements (both in translation and distance) and presented as the average of the locations of last several frames.

If you have any question, feel free to reach me at [email protected]. Welcome to checkout my homepage.

This work was done during my undergraduate internship at Institute of Automation, Chinese Academy of Sciences.

Dependencies:

Boost, Dlib 18.10, OpenCV 2.4, GLUT.

Note: Please make sure conn.txt and landmark.txt and dlib's default face landmarking model file shape_predictor_68_face_landmarks.dat (download here) are in the same directory as the executable.

Installation:

mkdir build

cd build

cmake .. -Ddlib_DIR=path/to/dlib/ (make sure the file dlib_DIR/dlib/cmake exists)

make

Usage:

the program will use your webcamera by default.

You can specify a video file as the first command line argument.

Acknowledgement:

The inspiration comes from IntraFace.

  • Jose Gonzalez-Mora, Fernando De la Torreb, Nicolas Guil, Emilio L.Zapata, 2010

    Learning a generic 3D face model from 2D image databases using incremental structure from motion

  • Vahid Kazemi and Josephine Sullivan

    One Millisecond Face Alignment with an Ensemble of Regression Trees

  • Lorenzo Torresani, Aaron Hertzmann and Christoph Bregler, NIPS 16, 2003

    Learning Non-Rigid 3D Shape from 2D Motion

  • Gianluca Donato and Serge Belongie, 2002

    Approximation Methods for Thin Plate Spline Mappings and Principal Warps

License:

Copyright (C) 2016 by Rui Huang [email protected]

The program DenseFace3D is licensed under the GNU General Public License.

DenseFace3D is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.

DenseFace3D is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

You should have received a copy of the GNU General Public License along with DenseFace3D. If not, see http://www.gnu.org/licenses/.

About

a software for the reconstruction of a real-time 3D face model from a monocular camera.

Resources

Readme
Activity

Stars

91 stars

Watchers

9 watching

Forks

48 forks
Report repository

Releases

Packages

Contributors

Languages

You are visitor 01402503 — GeezerVison does not store the pages you look up.
Best viewed with Netscape Navigator 4.7 or Internet Explorer 6 at 800×600. No cookies required.