Itseez engineers have co-authored a paper about running computer vision algorithms on GPU devices. We have covered both desktop GPU and mobile, addressing both performance and power issues. The paper is published by ACM Queue http://queue.acm.org/detail.cfm?id=2206309.

Itseez team has released OpenCV 2.4! We put a lot of efforts to make it one of the most stable OpenCV releases, over 300 issues have been resolved in the past few months. Among new features are: significantly improved support for Android, including Android x86, optical flow algorithms for GPU, Line-MOD detector of textureless objects and many more! For the full list of changes, please see http://code.opencv.org/projects/opencv/wiki/ChangeLog. 

Itseez machine learning contest in Nizhny State University has completed! See the winners on the contest site!

The Khronos group has announced the new initiative to create an open, royalty-free standard for cross platform acceleration of computer vision applications. Itseez as a company that proposed the creation of the computer vision working group has become a Khronos member and will take active part in the creation of the standard. This is an important step toward our goal of providing development tools for creating efficient computer vision applications! For more information, please see the Khronos press release.

We are happy to announce the development of a new open source implementation of KinectFusion, a simple system for 3D local mapping with an OpenNI-compatible camera. A description of the algorithm is presented in KinectFusion: Real-time 3D Reconstruction and Interaction Using a Moving Depth Camera, and you can watch a brief demonstration of the current implementation's capabilities in the video. The preliminary version of the code is released open source as a part of the PCL library. See more details at pointclouds.org.