people
projects
activities
publications

CVIP is a research theme within the VIVE research group. Computer vision and image processing aim to extract useful information from images and movies. Possible applications include face recognition, automatic construction of image databases, robotics, visual inspection for production lines, building 3D models of the real world etc. etc.

One of the reasons that computer vision is an interesting research problem is the sheer amount of data involved. A VGA (640x480) camera collecting data at 30Hz provides 27 Mb of image data per second. One interesting issue in computer vision is how to efficiently extract salient parts of this huge datastream. Another research area has investigated the geometry of multiple cameras and how they relate to objects in the scene. A further area of work involves building statistical models of the appearance and shape of objects in the scene. For example, these can be used for object detection.

Computer vision and image processing make use of mathematical techniques including geometry, statistics, physics, statistical decision theory, signal processing, algorithmics and analysis/partial differential equations. Computer vision is closely related to several other academic fields, including computer graphics, neural networks and machine learning, robotics, biological vision and wearable computing.

There are a number of researchers in UCL computer science with interests in computer vision and image processing. Particular interests include face recognition, statistical modelling of colour information, tracking for augmented reality, image based rendering and building mathematical models of shape and appearance.