Research Fellow in Computer Vision and Deep Learning

Страна: Великобритания;

Дедлайн: 12.06.2019

Тип гранта: На научные исследования;


Область наук: Компьютерные;


The Centre for Vision, Speech, and Signal Processing is seeking to recruit two full-time postdoctoral researchers to contribute to research projects.

The EPSRC project ROSSINI (Reconstructing 3D structure from single images: a perceptual reconstruction approach) will develop deep learning solutions to estimating 3D structure from 2D images and video. This project is in collaboration with partners in Psychology at the Universities of Aston and the University of Southampton. The project seeks to better understand how humans perceive 3D cues in 2D images and how this can factor into computer vision and deep learning approaches. 

The EPSRC project ExTOL (End to End Translation of British Sign Language) will develop deep learning solutions for human pose estimation, facial feature tracking, and ultimately language translation. This project is in close collaboration with the Visual Geometry Group at the University of Oxford and the Deafness, Cognition and Language Research Centre at UCL.

We also have several projects with close industrial collaboration on perception and decision making for autonomous vehicles. This work builds on the Surrey autonomous testbed so there are further opportunities to contribute to this work and deploy your code on a real autonomous vehicle.

You will join an established research group (165 researchers) within a highly rated research department, with an excellent computing and software infrastructure (GPU compute) for first class research, modern facilities and close links with industry.

Applicants should have (or be in the final stages of completing) a PhD in a related area with a track record of publishing high quality research at conferences such as CVPR, ICCV, ECCV or BMVC. You should have good analytical and programming skills and experience of computer vision and machine/deep learning. Prior experience of deep learning in either pyTorch or TensorFlow is essential, ROS expertise would be beneficial. Applicants are expected to be proactive, to support other researchers and to be self-motivated.

Informal enquiries related to the project to Professor Richard Bowden by e-mail: . 

Please note, it is University Policy to offer a starting salary equivalent to Level 3.6 (£31,302) to successful applicants who have been awarded, but are yet to receive, their PhD certificate. Once the original PhD certificate has been submitted to the local HR Department, the salary will be increased to Level 4.1 (£32,236).