Git repository ^_^

YAHOO! We have Git repository at code.google. This project consists of three branch: debugger, library, logger and release - master. Each developer from team push his commits to some branch. And when we want tar realese - merge all branches => Master.
Everybody can folow the development of the project.

Link to the our project: CV-T-PRO

Useful link

Here is a useful link to publications of the Visual Geometry Group (the part of Robotics Research Group in the Department of Engineering Science, University of Oxford): VGG publications

Searching of erroneous correspondences of key points

  We implemented two algorithms which searches  erroneous correspondences of key points. 

The 1^st algorithm.

Key points are grouped using the  recursive algorithm breadth-first search. Then for every group calculated average slope angle of the vector points and the average length of a vector. Vectors, whose parameters differ significantly from the average parameters identifies as erroneous.

This algorithm is invariant to rotation and scaling.

Cvt Debuger 1.1.3

Hi
We have news about cvt debuger:

1. Fixed known bugs.

2. Fluid Layout Widgets

3. Added the ability to capture video from connected cameras

4. Added a widget for changing speed video

In the next release:

1. Include vision library.
2. Create a widgets for configuration vision library.

~fix opencv_2 bug with playing video.

Asus, PrimeSense Reveals Motion Sensing for PC

It's a Kinect-like device for the PC, however it's by ASUS and called WAVI Xtion.

Both ASUS and PrimeSense announced a joint project that will bring motion sensing controls to the PC, thus possibly debunking any rumors that Microsoft will eventually bring Windows-based support for the Xbox 360's Kinect motion sensing device.

According to both companies, the WAVI Xtion for the PC will utilize PrimeSense's core 3D sensing solution already licensed by Microsoft for Kinect. The difference will be that the PC-oriented device will be developed primarily for browsing multimedia content, accessing Web sites and social networks when the PC is hooked up to the living room TV.

ZCam

ZCam is a brand of time-of-flight camera products for video applications by Israeli developer 3DV Systems. The ZCam supplements full-color video camera imaging with real-time range imaginginformation, allowing for the capture of video in 3D.

The original ZCam, released in 2000, was an ENG video camera add-on used for digital video compositing. Before agreeing in March 2009 to sell its assets to Microsoft, 3DV had planned to release a ranging video webcam (previously called the Z-Sense), also under the name ZCam. The ZCam webcam was one of several competing real-time range imaging camera products in development that target home game controller applications.

Technology.

The ZCam's time-of-flight camera system features a near-infrared (NIR) pulse illumination component, as well as an image sensor with a fast gating mechanism. Based on the known speed of light, ZCam coordinates the timing of NIR pulse wave emissions from the illuminator with the gating of the image sensor, so that the signal reflected from within a desired depth range is captured exclusively. The amount of pulse signal collected for each pixel corresponds to where within the depth range the pulse was reflected from, and can thus be used to calculate the distance to a corresponding point on the captured subject.

OpenCV Camera Calibration

Camera calibration is important for any image processing to be a highly accurate representation of the real world. "Geometric entities of the camera, such as the projection centre and image plane, can be computed quite simply from its matrix representation. Specializations of the general projective camera inherit its properties, for example their geometry is computed using the same algebraic expressions."

This algorithm show how to extract the relevant camera parameters from a sequence of images with using of OpenCV library. For this example you will need to print out this image as a full page:

Example Checkerboard

3D CV-t debuger :)

We are in the process of developing application for testing implementations of recognition methods

The development plan:
1) Implement the ability to view video in 3D scene. [done]
2) The free movement in a 3d scene. [done]
3) The process of video playback [done]
4) User can load functions (software modules) for detection (which are used for video playback). [in process]

KinectFusion: Real-Time Dense Surface Mapping and Tracking

Kinect is a motion sensing input device by Microsoft for the Xbox 360 video game console. Based around a webcam-style add-on peripheral for the Xbox 360 console, it enables users to control and interact with the Xbox 360 without the need to touch a game controller, through a natural user interface using gestures and spoken commands.

Microsoft KinectFusion software provides real-time to recover 3D scene based on data from the Kinect to perform segmentation and tracking of objects.

Searching pairs of identical key points

The problem of searching pairs of identical key points on two images has been solved with using of algorithm that performs a simple search and comparison of elements of descriptors arrays.

The same problem has been realized with using of FLANN library in OpenCV.

Congratulations!

We congratulate the Odessa office of the Luxoft with 5th anniversary! The company has arranged a very good entertainment for your program's staff - drift show and corporate event. Here are some photos from the event:
We wish the company's growth and prosperity in all her endeavors! We are proud that we are working with such wonderful people!
Thank you for reading our blog!

The long-awaited meeting!

Vice President Eastern Europe at Luxoft Dmitry Kushnir has visited us today!

 We have shown him the results our our work. We tried very hard to show him, that we are really good programmers.
We hope, he really like it!

Training for students!

Now we are training our students. We read lectures about the syntax of programming languages such as C++, the correctness of the construction of algorithms and the structure of Linux-based systems. We hope that this will be useful to us and them in their future work!
The first training will be held on the construction of algorithms on September 22. The second training in Linux will be next week, stay tuned!
Thank you to join us!

CW37 summary report

This week we have been able to write a program that takes video from a file and it can rotate in three dimensions by changing the coordinate axes. It called "Frame grabber to 3D". It is not yet completed the program but we already have the opportunity to present it to you:

We will continue developing this program in the future and we will make it more useful for our purposes. We will inform you regularly on the interim results of our development. Thank you to join us!



CW36 report: Team completion.

This week, we selected the other students, who will subsequently work with us and take our place in the future developments. The full list is presented here:
 Navit-team: Vladislav Dramaretskiy, Ievhen Kolusetskiy and Ivan Nazarenko.
 Augmented-team: Andrey Mazein, Alexey Odinochenko and Asya Arutyunyan.

They will work with us on October 1. We will be happy to see them in our teams!

CW35 summing-up by Augmented-team

After a week of work we managed to develop a program to recognize individuals in video stream from an ordinary web camera. Here are some screenshots showing the program results in real time:

The result was really stunning, and despite the complexity of the task was completed in a short time. The program was written in a programming language C++ and uses openCV. OpenCV (Open Source Computer Vision Library) is a library of programming functions mainly aimed at real time computer vision. Thus we first met with computer vision, and we really became interesting. We hope that you will support us in the future.

We are glad to see you on our blog!

Hello, we are command of young developers, code-named Crocod1le. We are glad to see you on the pages of our blog!

Our group divided into teams that solve different problems:

The first team solve problems with modern navigation systems. They're trying to make plugin for Navit. Navit-team: Arthur Fayt, Dmitriy Semenakha and Oksana Grygoryeva.

Another team carries out the tasks of pattern recognition. The main purpose is to develop on-board computer program which is able to recognize patterns of pedestrians, traffic signs, road markings, to make the car drive itself wherever you want! The work is just beginning and much remains to be done. They hope that their efforts will not be ignored!
Augmented-team: Artem Gayduchenko, Ivan Siyanko and Ievhen Okhrimenko.

We hope our blog would be interesting for you! Thank you!