xvWidgets
xvWidgets Documentation

An experimental cross-platform C++ GUI toolkit for computer vision applications with (planned) Qt and wxWidgets bindings

Introduction

The goal of the project is to consolidate the code base and lessons learned from developing desktop applications, with both computer vision and GUI requirements, into a set of reusable components. The initials xv stand for an arbitrary GUI toolkit (x) and vision (v).

Status

The library is in an early development phase, major API and architecture changes can happen at any time. It should become minimally usable and stable from the moment sample applications appear in the repository. The initial OpenCV centric architecture is being dropped in favor of an approach that doesn't depend on any particular CV library. OpenCV is still supported and most sample code will be based on it.

Source Code

The source code is hosted in GitHub.

Documentation

The Doxygen generated API reference is hosted in oss.witstep.com.

Usage

The raison d'etre of the library was the need for a simple and intuitive way of accepting user input for basic 2D shapes, as exemplified bellow.

Polygon Input

Clip from Tears of Steel | (CC) Blender Foundation | mango.blender.org

Modules