{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,2]],"date-time":"2026-05-02T10:05:34Z","timestamp":1777716334959,"version":"3.51.4"},"reference-count":49,"publisher":"SAGE Publications","issue":"5","license":[{"start":{"date-parts":[[2013,4,1]],"date-time":"2013-04-01T00:00:00Z","timestamp":1364774400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/journals.sagepub.com\/page\/policies\/text-and-data-mining-license"}],"content-domain":{"domain":["journals.sagepub.com"],"crossmark-restriction":true},"short-container-title":["The International Journal of Robotics Research"],"published-print":{"date-parts":[[2013,4]]},"abstract":"<jats:p>This paper addresses the visual odometry problem from a machine learning perspective. Optical flow information from a single camera is used as input for a multiple-output Gaussian process (MOGP) framework, that estimates linear and angular camera velocities. This approach has several benefits. (1) It substitutes the need for conventional camera calibration, by introducing a semi-parametric model that is able to capture nuances that a strictly parametric geometric model struggles with. (2) It is able to recover absolute scale if a range sensor (e.g. a laser scanner) is used for ground-truth, provided that training and testing data share a certain similarity. (3) It is naturally able to provide measurement uncertainties. We extend the standard MOGP framework to include the ability to infer joint estimates (full covariance matrices) for both translation and rotation, taking advantage of the fact that all estimates are correlated since they are derived from the same vehicle. We also modify the common zero mean assumption of a Gaussian process to accommodate a standard geometric model of the camera, thus providing an initial estimate that is then further refined by the non-parametric model. Both Gaussian process hyperparameters and camera parameters are trained simultaneously, so there is still no need for traditional camera calibration, although if these values are known they can be used to speed up training. This approach has been tested in a wide variety of situations, both 2D in urban and off-road environments (two degrees of freedom) and 3D with unmanned aerial vehicles (six degrees of freedom), with results that are comparable to standard state-of-the-art visual odometry algorithms and even more traditional methods, such as wheel encoders and laser-based Iterative Closest Point. We also test its limits to generalize over environment changes by varying training and testing conditions independently, and also by changing cameras between training and testing.<\/jats:p>","DOI":"10.1177\/0278364912472245","type":"journal-article","created":{"date-parts":[[2013,5,17]],"date-time":"2013-05-17T11:08:23Z","timestamp":1368788903000},"page":"526-546","update-policy":"https:\/\/doi.org\/10.1177\/sage-journals-update-policy","source":"Crossref","is-referenced-by-count":17,"title":["Semi-parametric learning for visual odometry"],"prefix":"10.1177","volume":"32","author":[{"given":"Vitor","family":"Guizilini","sequence":"first","affiliation":[{"name":"Australian Centre for Field Robotics, School of Information Technologies, University of Sydney, Australia"}]},{"given":"Fabio","family":"Ramos","sequence":"additional","affiliation":[{"name":"Australian Centre for Field Robotics, School of Information Technologies, University of Sydney, Australia"}]}],"member":"179","published-online":{"date-parts":[[2013,5,17]]},"reference":[{"key":"bibr1-0278364912472245","volume-title":"Proceedings of the International Conference on Advanced Robotics (ICAR)","author":"Agrawal M","year":"2007"},{"key":"bibr2-0278364912472245","doi-asserted-by":"publisher","DOI":"10.1007\/11744023_32"},{"key":"bibr3-0278364912472245","first-page":"153","volume-title":"Proceedings of Advances in Neural Information Processing Systems (NIPS)","author":"Bonilla EV","year":"2008"},{"key":"bibr4-0278364912472245","volume-title":"Proceedings of the 6th Latin American Robotics Symposium (LARS)","author":"Botelho S","year":"2009"},{"key":"bibr5-0278364912472245","series-title":"Technical report","volume-title":"Multiple Output Gaussian Process Regression","author":"Boyle P","year":"2005"},{"key":"bibr6-0278364912472245","doi-asserted-by":"publisher","DOI":"10.1109\/IROS.2004.1389991"},{"key":"bibr7-0278364912472245","volume-title":"Proceedings of Advances in Neural Information Processing Systems (NIPS)","author":"Chai KM","year":"2008"},{"key":"bibr8-0278364912472245","doi-asserted-by":"publisher","DOI":"10.1109\/ICSMC.2005.1571261"},{"key":"bibr9-0278364912472245","doi-asserted-by":"publisher","DOI":"10.1109\/ROBOT.2007.364181"},{"key":"bibr10-0278364912472245","doi-asserted-by":"publisher","DOI":"10.1109\/IROS.2004.1390041"},{"key":"bibr11-0278364912472245","doi-asserted-by":"publisher","DOI":"10.1002\/9781119115151"},{"key":"bibr12-0278364912472245","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2003.1238654"},{"key":"bibr13-0278364912472245","series-title":"Technical report","volume-title":"The Expectation Maximization Algorithm","author":"Dellaert F","year":"2002"},{"key":"bibr14-0278364912472245","doi-asserted-by":"publisher","DOI":"10.1145\/358669.358692"},{"key":"bibr15-0278364912472245","volume-title":"12th International Symposium on Experimental Robotics (ISER)","author":"Guizilini V","year":"2010"},{"key":"bibr16-0278364912472245","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA.2012.6224775"},{"key":"bibr17-0278364912472245","doi-asserted-by":"publisher","DOI":"10.1017\/CBO9780511811685"},{"key":"bibr18-0278364912472245","doi-asserted-by":"publisher","DOI":"10.1007\/978-0-387-21606-5"},{"key":"bibr19-0278364912472245","doi-asserted-by":"publisher","DOI":"10.1007\/978-1-4471-0657-9_2"},{"key":"bibr20-0278364912472245","doi-asserted-by":"publisher","DOI":"10.1109\/IROS.2008.4651147"},{"key":"bibr21-0278364912472245","volume-title":"15th International Symposium on Robotics Research (ISRR)","author":"Huang AS","year":"2011"},{"key":"bibr22-0278364912472245","volume-title":"Proceedings of the 6th International Conference on Field and Service Robotics","author":"Kelly J","year":"2007"},{"key":"bibr23-0278364912472245","volume-title":"Proceedings of the 6th IFAC Symposium on Intelligent Autonomous Vehicles","author":"Kelly J","year":"2007"},{"key":"bibr24-0278364912472245","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-007-0042-3"},{"key":"bibr25-0278364912472245","doi-asserted-by":"publisher","DOI":"10.1109\/IVS.2011.5940546"},{"key":"bibr26-0278364912472245","doi-asserted-by":"publisher","DOI":"10.1023\/B:VISI.0000029664.99615.94"},{"key":"bibr27-0278364912472245","volume-title":"IEEE Computer Vision and Pattern Recognition Conference (CVPR)","author":"Lu F","year":"1994"},{"key":"bibr28-0278364912472245","volume-title":"Proceedings of the DARPA Image Understanding Workshop","author":"Lucas BD","year":"1981"},{"key":"bibr29-0278364912472245","volume-title":"Information Theory, Inference and Learning Algorithms","author":"MacKsay DJC","year":"2002"},{"key":"bibr30-0278364912472245","series-title":"Technical report","volume-title":"Cart Project Progress Report","author":"Moravec H","year":"1976"},{"key":"bibr31-0278364912472245","volume-title":"Obstacle Avoidance and Navigation in the Real World by a Seeing Robot Rover","author":"Moravec HP","year":"1980"},{"key":"bibr32-0278364912472245","doi-asserted-by":"publisher","DOI":"10.1007\/978-1-4612-0745-0"},{"key":"bibr33-0278364912472245","doi-asserted-by":"publisher","DOI":"10.1002\/rob.20103"},{"key":"bibr34-0278364912472245","doi-asserted-by":"publisher","DOI":"10.1109\/ROBOT.2009.5152754"},{"key":"bibr35-0278364912472245","volume-title":"Gaussian Processes for Machine Learning","author":"Rasmussen CE","year":"2006"},{"key":"bibr36-0278364912472245","doi-asserted-by":"publisher","DOI":"10.1109\/ROBOT.2008.4543185"},{"key":"bibr37-0278364912472245","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206538"},{"key":"bibr38-0278364912472245","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2009.5459294"},{"key":"bibr39-0278364912472245","doi-asserted-by":"publisher","DOI":"10.1109\/TRO.2008.2004490"},{"key":"bibr40-0278364912472245","doi-asserted-by":"publisher","DOI":"10.1109\/ROBOT.2001.932909"},{"key":"bibr41-0278364912472245","volume-title":"Visual Odometry using Sparse Bundle Adjustment on an Autonomous Outdoor Vehicle","author":"Sunderhauf N","year":"2005"},{"key":"bibr42-0278364912472245","doi-asserted-by":"publisher","DOI":"10.1109\/IROS.2008.4651205"},{"key":"bibr43-0278364912472245","doi-asserted-by":"publisher","DOI":"10.1111\/1467-9868.00196"},{"key":"bibr44-0278364912472245","first-page":"391","volume-title":"Proc. 7th International Symposium on Robotics Research (ISRR)","author":"Tomasi C","year":"1995"},{"key":"bibr45-0278364912472245","volume-title":"IEEE Computer Vision and Pattern Recognition Conference (CVPR)","author":"Tomasi S","year":"1994"},{"key":"bibr46-0278364912472245","doi-asserted-by":"publisher","DOI":"10.1109\/ROBOT.2009.5152677"},{"key":"bibr47-0278364912472245","doi-asserted-by":"publisher","DOI":"10.1177\/0278364906075026"},{"key":"bibr48-0278364912472245","doi-asserted-by":"publisher","DOI":"10.1162\/089976698300017412"},{"key":"bibr49-0278364912472245","volume-title":"Proceedings of the Workshop of Performance Metrics for Intelligent Systems","author":"Zhu ZW","year":"2006"}],"container-title":["The International Journal of Robotics Research"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/0278364912472245","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/full-xml\/10.1177\/0278364912472245","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/0278364912472245","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T10:18:04Z","timestamp":1777457884000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.sagepub.com\/doi\/10.1177\/0278364912472245"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2013,4]]},"references-count":49,"journal-issue":{"issue":"5","published-print":{"date-parts":[[2013,4]]}},"alternative-id":["10.1177\/0278364912472245"],"URL":"https:\/\/doi.org\/10.1177\/0278364912472245","relation":{},"ISSN":["0278-3649","1741-3176"],"issn-type":[{"value":"0278-3649","type":"print"},{"value":"1741-3176","type":"electronic"}],"subject":[],"published":{"date-parts":[[2013,4]]}}}