Subcategory-aware object detection

Xiaoyuan Yu, Jianchao Yang, Zhe Lin, Jiangping Wang, Tianjiang Wang, Thomas Huang

Research output: Contribution to journalArticlepeer-review

Abstract

In this letter, we introduce a subcategory-aware object detection framework to detect generic object classes with high intra-class variace. Motivated by the observation that the object appearance demonstrates some clustering property, we split the training data into subcategories and train a detector for each subcategory. Since the proposed ensemble of detectors relies heavily on subcategory clustering, we propose an effective subcategories generation method that is tuned for the detection task. More specifically, we first initialize subcategories by constrained spectral clustering based on mid-level image features used in object recognition. Then we jointly learn the ensemble detectors and the latent subcategories in an alternative manner. Our performance on the PASCAL VOC 2007 detection challenges and INRIA Person dataset is comparable with state-of-the-art, even with much less computational cost.

Original languageEnglish (US)
Article number6709751
Pages (from-to)1472-1476
Number of pages5
JournalIEEE Signal Processing Letters
Volume22
Issue number9
DOIs
StatePublished - Sep 1 2015

Keywords

  • Constrained spectral cluttering
  • joint subcategories learning
  • max pooling
  • object detection
  • subcategory-aware

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering
  • Applied Mathematics

Fingerprint Dive into the research topics of 'Subcategory-aware object detection'. Together they form a unique fingerprint.

Cite this