Skip to main content

QtWS15- Qt, multimedia, and computer vision – a great match for Embedded Linux, Laszlo Agocs

Qt World Summit 2015, OCTOBER 5-7, BERLIN, GERMANY Qt, multimedia, and computer vision – a great match for Embedded Linux: In this live coding session we are going to build a cross-platform computer vision application utilizing the camera. Based on Qt Quick Controls, Qt Multimedia, and OpenCV, the application will run both on desktop platforms and Embedded Linux. The application will be developed in Qt Creator and deployed onto an i.MX6-based board running the software stack provided by Qt for Device Creation. Qt Multimedia makes creating Qt Quick applications featuring accelerated video and camera streams as a natural part of their user interface quick and painless. The new video filtering features of Qt 5.5 further enhance these capabilities: integrating 3rd party vision and image processing frameworks, like OpenCV, or accelerated algorithms implemented with GPU compute APIs, like OpenCL and CUDA, is now made simple and efficient. OpenCV (Open Source Computer Vision Library) a BSD-licensed computer vision and machine learning software library with more than 2500 optimized algorithms. Application areas include object detection, motion tracking and estimation, facial and gesture recognition, machine learning, and others. Computer vision is an area of continuously increasing focus in the mobile and embedded world. Its role is central in automotive, robotics, and augmented reality. Being a C++ framework, Qt allows easy and straightforward usage of external libraries like OpenCV. Combined with the touch-friendly user interface elements of Qt Quick Controls and the Flat style, Qt becomes an excellent choice for applications and embedded products utilizing computer vision algorithms. It is now time to see all this in action.