Visual computing
Visual Computing is an umbrella term for all computer science disciplines that deal with image information and 3D models.
These include computer graphics , image processing , visualization , computer vision , virtual reality and augmented reality , video processing , but also aspects of pattern recognition , human-computer interaction , machine learning and digital libraries . This involves the acquisition, processing, analysis and presentation of visual information (mainly images and videos).
Areas of application are e.g. B. industrial quality control , medical image processing and visualization, surveying , robotics , multimedia systems , virtual heritage , visual effects in film and television, and computer games .
Several German-speaking universities offer their own courses in visual computing .
History and overview
Visual computing is still a relatively new term, which only gained its current meaning in 2005, when the disciplines computer graphics , image processing , computer vision and the like were established in information technology . Ä. in their methodology and their applications moved closer together and a generic term was needed. Many of the mathematical and algorithmic methods used are the same in all areas that use images: image formats , filtering methods , color models , image metrics, and others. The programming methods on graphics hardware, the processing of large amounts of data , the textbooks and the congresses, the scientific communities of these disciplines and the working groups in companies are becoming more and more mixed up.
In addition, applications increasingly require techniques from several of these disciplines at the same time. In order to generate detailed models of complex objects, image recognition, 3D sensors and reconstruction processes are required. Realistic rendering processes with complex lighting simulation are required to represent these models in a credible manner. Real-time graphics are the basis for usable virtual and augmented reality software. A good segmentation of the organs is the basis for interactive manipulations in the 3D representation of medical scans. Robot control requires a recognition of the objects as well as a modeling of the environment. And an ergonomic graphic interface with the devices used is also necessary.
Although many tasks in the sub-disciplines are considered to be solved in the scientific world (mostly under "laboratory conditions"), an essential task of the overall visual computing discipline is the integration of the sub-solutions into usable products - this also includes the treatment of many practical problems, from dealing with Hardware variety through the use of real (mostly faulty or extensive) data to operation by untrained users.
Areas of visual computing
At least the following areas belong to visual computing. Detailed descriptions of each of these areas can be found on the respective special pages for the terms.
Computer graphics and computer animation
Computer graphics is a generic term for all techniques that generate image information as a result of a calculation process. With image synthesis (rendering) images are generated from descriptions of objects which, depending on the application, usually represent a compromise between quality and computing time. Computer animation is called computer graphics when the images are generated for the purpose of creating a film.
Image recognition and computer vision
Techniques that can extract information about the content from existing images fall under the term image recognition . Under Computer vision is the ability of the computer (or robot) to recognize the surroundings and interpret correctly.
Visualization and Interactive Visual Analysis
The term visualization is mostly used when data that for whatever reason is not directly accessible are presented as clearly as possible. This applies in particular to volume data and data that have no direct geometric dimension. With interactive visual analysis , you can efficiently examine confusing amounts of data by changing the display interactively.
Geometric modeling and 3D printer
The representation of representable objects in the computer requires special methods and data structures that are known under the term geometric modeling . In addition to descriptive and interactive geometric techniques, sensor data are also increasingly being used to reconstruct the geometric models. More recently, algorithms for the efficient control of 3D printers have also been counted as visual computing.
Image processing and image editing
In contrast to image recognition, image processing is used to calculate better images from images. “Better” can have very different meanings depending on the application. A distinction must be made between image processing , which deals with interactive methods for manually changing images.
Virtual and augmented reality as well as sensor technology
Virtual Reality is the name of all techniques that suggest to the user that they are in a fictional environment. In addition to good data glasses , you need precise tracking and high-quality real-time graphics . With augmented reality , you can also see the real environment, which increases the demands on the accuracy of the visual representation and localization considerably.
Graphic components of human-machine interaction
The human-computer interaction (or English Human Computer Interaction ) said region deals with the user-oriented design of interactive systems, where the graphical components are of particular importance because of the visual channel of the people has the highest bandwidth for receiving information.