Please use this identifier to cite or link to this item:
Scopus Web of Science® Altmetric
Type: Conference paper
Title: Approximate constraint generation for efficient structured boosting
Author: Lin, G.
Shen, C.
Van Den Hengel, A.
Citation: 2013 IEEE International Conference on Image Processing, ICIP 2013 Proceedings, Melbourne, Vic., Australia, 15-18 September 2013: pp.4287-4291
Publisher: IEEE
Publisher Place: USA
Issue Date: 2013
Series/Report no.: IEEE International Conference on Image Processing ICIP
ISBN: 9781479923410
ISSN: 1522-4880
Conference Name: International Conference on Image Processing (20th : 2013 : Melbourne, Australia)
Statement of
Guosheng Lin, Chunhua Shen, Anton van den Hengel
Abstract: We propose efficient training methods (SBoost) for totally-corrective boosting based structured learning. The optimization of boosting method for structured learning is more challenging than the structured support vector machine. Basically, we propose smooth and convex formulation for boosting based structured learning, and develop approximate constraint generation together with column generation to solve the optimization with large number of constraints and variables. Because of the convexity and smoothness, the optimization in each generation iteration can be solved efficiently. We demonstrate some structured learning applications in computer vision using SBoost, including invariance learning for digit recognition, object detection and hierarchical image classification.
Rights: © 2013 IEEE
RMID: 0020137301
DOI: 10.1109/ICIP.2013.6738883
Appears in Collections:Computer Science publications

Files in This Item:
There are no files associated with this item.

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.