Burke, Michael G2017-06-072017-06-072016-12Burke, M. 2016. Image ranking in video sequences using pairwise image comparisons and temporal smoothing. 2016 Pattern Recognition Association of South Africa and Robotics and Mechatronics International Conference (PRASA-RobMech), 30 November-2 December 2016, Stellenbosch, Cape Town. DOI: 10.1109/RoboMech.2016.7813166978-1-5090-3335-5DOI: 10.1109/RoboMech.2016.7813166http://ieeexplore.ieee.org/document/7813166/http://hdl.handle.net/10204/91742016 Pattern Recognition Association of South Africa and Robotics and Mechatronics International Conference (PRASA-RobMech), 30 November-2 December 2016, Stellenbosch, Cape Town.The ability to predict the importance of an image is highly desirable in computer vision. This work introduces an image ranking scheme suitable for use in video or image sequences. Pairwise image comparisons are used to determine image ‘interest’ values within a standard Bayesian ranking framework, and a Rauch-Tung-Striebel smoother is used to improve these interest scores. Results show that the training data requirements typically associated with pairwise ranking systems are dramatically reduced by incorporating temporal smoothness constraints. Experiments on a coastal image dataset show that smoothed pairwise ranking can provide ranking results equivalent to standard pairwise ranking with less than half the training data.enImage rankingBayesian modellingInterest detectionConference PresentationBurke, M. G. (2016). Image ranking in video sequences using pairwise image comparisons and temporal smoothing. IEEE. http://hdl.handle.net/10204/9174Burke, Michael G. "Image ranking in video sequences using pairwise image comparisons and temporal smoothing." (2016): http://hdl.handle.net/10204/9174Burke MG, Image ranking in video sequences using pairwise image comparisons and temporal smoothing; IEEE; 2016. http://hdl.handle.net/10204/9174 .TY - Conference Presentation AU - Burke, Michael G AB - The ability to predict the importance of an image is highly desirable in computer vision. This work introduces an image ranking scheme suitable for use in video or image sequences. Pairwise image comparisons are used to determine image ‘interest’ values within a standard Bayesian ranking framework, and a Rauch-Tung-Striebel smoother is used to improve these interest scores. Results show that the training data requirements typically associated with pairwise ranking systems are dramatically reduced by incorporating temporal smoothness constraints. Experiments on a coastal image dataset show that smoothed pairwise ranking can provide ranking results equivalent to standard pairwise ranking with less than half the training data. DA - 2016-12 DB - ResearchSpace DP - CSIR KW - Image ranking KW - Bayesian modelling KW - Interest detection LK - https://researchspace.csir.co.za PY - 2016 SM - 978-1-5090-3335-5 T1 - Image ranking in video sequences using pairwise image comparisons and temporal smoothing TI - Image ranking in video sequences using pairwise image comparisons and temporal smoothing UR - http://hdl.handle.net/10204/9174 ER -