Home
Welcome to the openMVG wiki!
You will find multiple resources to discover OpenMVG and how to use it for your own purpose.
- OpenMVG is a library mainly focused on Multiple-View-Geometry and Structure-From-Motion.
- Those computer vision tasks require to find corresponding points between images and the manipulation of 3D geometry to define the relationship between 3D points and their projection as 2D features in images thanks to the definition of camera models.
- Obtaining corresponding points in an image collection is using the computer vision tasks known as
Image Description,Image Retrieval/Matching,Feature Tracking, ..., then you can solve the motion of the camera by solving the 3D structure corresponding to the triangulation of the 2d images observations.
OpenMVG provides an end-to-end 3D reconstruction from images framework compounded of 3 layers:
- Libraries contains the core logic implementation and algorithms implementation that anyone can reuse in his code.
- Binaries provides easy access to important steps necessary to build a pipeline.
- Pipelines that chain multiple binaries to solve structure from motion problems starting from images.
In order to let you discover OpenMVG we invite you to read the following sections: