Learning to Group Objects

V. Yanulevskaya, J. R. R. Uijlings, N. Sebe

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


This paper presents a novel method to generate a hypothesis set of class-independent object regions. It has been shown that such object regions can be used to focus computer vision techniques on the parts of an image that matter most leading to significant improvements in both object localisation and semantic segmentation in recent years. Of course, the higher quality of class-independent object regions, the better subsequent computer vision algorithms can perform. In this paper we focus on generating higher quality object hypotheses. We start from an oversegmentation for which we propose to extract a wide variety of region-features. We group regions together in a hierarchical fashion, for which we train a Random Forest which predicts at each stage of the hierarchy the best possible merge. Hence unlike other approaches, we use relatively powerful features and classifiers at an early stage of the generation of likely object regions. Finally, we identify and combine stable regions in order to capture objects which consist of dissimilar parts. We show on the PASCAL 2007 and 2012 datasets that our method yields higher quality regions than competing approaches while it is at the same time more computationally efficient.
Original languageEnglish
Title of host publicationComputer Vision and Pattern Recognition (CVPR), 2014 IEEE Conference on
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
Number of pages8
ISBN (Print)978-1-4799-5117-8
Publication statusPublished - 1 Jun 2014


  • computer vision
  • feature extraction
  • image segmentation
  • learning (artificial intelligence)
  • object detection
  • trees (mathematics)
  • PASCAL dataset
  • class-independent object regions
  • computer vision algorithms
  • computer vision technique
  • higher quality object hypotheses generation
  • image parts
  • learning
  • object grouping
  • object localisation
  • oversegmentation
  • random forest training
  • region grouping
  • region-feature extraction
  • semantic segmentation
  • Computer vision
  • Feature extraction
  • Histograms
  • Image color analysis
  • Image segmentation
  • Merging
  • Radio frequency
  • Class independent object proposals
  • Selective Search
  • regions
  • segmentation


Dive into the research topics of 'Learning to Group Objects'. Together they form a unique fingerprint.

Cite this