University of Hertfordshire

Joint optimisation for object class segmentation and dense stereo reconstruction

Research output: Chapter in Book/Report/Conference proceedingConference contribution


  • Paper104

    Final published version, 368 KB, PDF document

  • L. Ladicky
  • S. Sengupta
  • C. Russell
  • P. Sturgess
  • Yalin Bastanlar
  • William Clocksin
  • P.H.S. Torr
View graph of relations
Original languageEnglish
Title of host publicationProceedings of the British Machine Vision Conference 2010
Subtitle of host publicationBMVC
PublisherBMVA Press
Pagespaper 104
Publication statusPublished - 2010
EventBritish Machine Vision Conf 2010 - Aberystwyth, United Kingdom
Duration: 31 Aug 20103 Sep 2010


ConferenceBritish Machine Vision Conf 2010
Country/TerritoryUnited Kingdom


The problems of dense stereo reconstruction and object class segmentation can both be formulated as Conditional Random Field based labelling problems, in which every pixel in the image is assigned a label corresponding to either its disparity, or an object class such as road or building. While these two problems are mutually informative, no attempt has been made to jointly optimise their labellings. In this work we provide a principled energy minimisation framework that unifies the two problems and demonstrate that, by resolving ambiguities in real world data, joint optimisation of the two problems substantially improves performance. To evaluate our method, we augment the street view
Leuven data set, producing 70 hand labelled object class and disparity maps. We hope that the release of these annotations will stimulate further work in the challenging domain of street-view analysis.


This work is supported by EPSRC research grants, HMGCC, TUBITAK researcher exchange grant, the IST Programme of the European Community, under the PASCAL2 Network of Excellence, IST-2007-216886.

ID: 502084