Supervised Feature Quantization with Entropy Optimization

Yubin Kuang, Martin Byröd, Karl Åström

Research output: Chapter in Book/Report/Conference proceedingPaper in conference proceedingpeer-review


Feature quantization is a crucial component for efficient large scale image retrieval and object recognition. By quantizing local features into visual words, one hopes that features that match each other obtain the same word ID. Then, similarities between images can be measured with respect to the corresponding histograms of visual words. Given the appearance variations of local features, traditional quantization methods do not take into account the distribution of matched features. In this paper, we investigate how to encode additional prior information on the feature distribution via entropy optimization by leveraging ground truth correspondence data. We propose a computationally efficient optimization scheme for large scale vocabulary training. The results from our experiments suggest that entropy-optimized vocabulary performs better than unsupervised quantization methods in terms of recall and precision for feature matching. We also demonstrate the advantage of the optimized vocabulary for image retrieval.
Original languageEnglish
Title of host publicationComputer Vision Workshops (ICCV Workshops), 2011 IEEE International Conference on
PublisherIEEE - Institute of Electrical and Electronics Engineers Inc.
Number of pages8
ISBN (Print)978-1-4673-0062-9 (print)
Publication statusPublished - 2011
Event1st IEEE Workshop on Information Theory in Computer Vision and Pattern Recognition (ICCV 2011), 2011 - Barcelona, Spain
Duration: 2011 Nov 62011 Nov 13


Conference1st IEEE Workshop on Information Theory in Computer Vision and Pattern Recognition (ICCV 2011), 2011

Subject classification (UKÄ)

  • Computer Vision and Robotics (Autonomous Systems)
  • Mathematics

Free keywords

  • visual vocabulary
  • entropy optimization


Dive into the research topics of 'Supervised Feature Quantization with Entropy Optimization'. Together they form a unique fingerprint.

Cite this