Home
Welcome to the wiki of "Eyes of Things", an international project focused on building the best embedded computer vision platform ever. EoT is an Innovation Action funded under the European Union’s H2020 Framework Programme that is going to unfold for three years starting in January 2015.
In this wiki are shown our main different projects with their implementations, in order to explain how they work. You can check their synopsis on this page. For more information, please choose any of them in the sidebar on the right.
The best way to contribute changes (e.g. a new feature, bug fixes, a great idea, etc.) to the project is through forks and pull request. All contributions are welcome and appreciated. Thanks for contributing!
OpenCV is a widely-used computer vision library with thousands of functions. OpenCV 1.0 has been ported to the EoT board. Additionally, functionality has been adder so that later OpenCV versions can be executed in an external server.
OpenCV is a computer vision library designed for computational efficiency and with a strong focus on real-time applications. Although OpenCV 1.0 has been ported to the EoT board, it is also important to have the possibility of running newer releases. To achieve this goal, an external server running an API to use OpenCV 2.4.3 is used. With this approach, it is actually possible to run any OpenCV version.
This application runs several examples using the OpenCV library.
[
] (https://www.flickr.com/photos/132007111@N03/25154115683/in/album-72157663571622754/)
Optical flow is the pattern of apparent motion on objects in a visual scene caused by the relative motion between an observer and a scene. The objective in this example is the application of Lucas-Kanade's method in the optical flow pattern.
[
] (https://www.flickr.com/photos/132007111@N03/25812548386/in/album-72157663571622754/)
libccv is a minimalist open-source computer vision library for embedded devices. It has been ported to the EoT board with some limitations.
This application runs several examples using the libccv library.
Quirc is a library for extracting and decoding QR codes from images. It is fast enough to be used with realtime video.
This application runs several examples using the Quirc library.
