Efficient discriminative local learning for object recognition

Yen-Yu Lin*, Jyun Fan Tsai, Tyng Luh Liu

*Corresponding author for this work

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

22 Scopus citations

Abstract

Although object recognition methods based on local learning can reasonably resolve the difficulties caused by the large variations in images from the same category, the high risk of overfitting and the heavy computational cost in training numerous local models (classifiers or distance functions) often limit their applicability. To address these two unpleasant issues, we cast the multiple, independent training processes of local models as a correlative multitask learning problem, and design a new boosting algorithm to accomplish it. Specifically, we establish a parametric space where these local models lie and spread as a manifold-like structure, and use boosting to perform local model training by completing the manifold embedding. Via sharing the common embedding space, the learning of each local model can be properly regularized by the extra knowledge from other models, while the training time is also significantly reduced. Experimental results on two benchmark datasets, Caltech-101 and VOC 2007, support that our approach not only achieves promising recognition rates but also gives a two order speed-up in realizing local learning.

Original languageEnglish
Title of host publication2009 IEEE 12th International Conference on Computer Vision, ICCV 2009
Pages598-605
Number of pages8
DOIs
StatePublished - 1 Dec 2009
Event12th International Conference on Computer Vision, ICCV 2009 - Kyoto, Japan
Duration: 29 Sep 20092 Oct 2009

Publication series

NameProceedings of the IEEE International Conference on Computer Vision

Conference

Conference12th International Conference on Computer Vision, ICCV 2009
CountryJapan
CityKyoto
Period29/09/092/10/09

Fingerprint Dive into the research topics of 'Efficient discriminative local learning for object recognition'. Together they form a unique fingerprint.

Cite this