Applications are invited for a Postdoc/RA in Computer Vision, focussing on visual recognition in dynamic outdoor scenes.
The aim of this project is to provide detailed and accurate recognition and segmentation of static and dynamic objects in video. Given video input from a moving camera, the aim will be to parse into objects/scene elements, given a learned library of static and dynamic elements. This will be used downstream as part of the Outdoor Asset Capture (OAK) project, to aid object-specific modelling of particular static/dynamic scene elements.
The ideal candidate will have a strong background in Computer Vision, and interest in object recognition and dynamic scene segmentation. Applicants must have a PhD degree in a relevant area, and previous experience and publications in areas relevant to the post would be highly desirable.