Please use this identifier to cite or link to this item:
Title: Incremental concept learning with few training examples and hierarchical classification
Authors: Bouma, Henri
Eendebak, Pieter T.
Schutte, Klamer
Azzopardi, George
Burghouts, Gertjan J.
Keywords: Sensor networks
Support vector machines
Image processing
Issue Date: 2015-09
Publisher: SPIE
Citation: Bouma, H., Eendebak, P. T., Schutte, K., Azzopardi, G., & Burghouts, G. J. (2015). Incremental concept learning with few training examples and hierarchical classification. SPIE - The International Society for Optical Engineering, Toulouse. 96520E-2.
Abstract: Object recognition and localization are important to automatically interpret video and allow better querying on its content. We propose a method for object localization that learns incrementally and addresses four key aspects. Firstly, we show that for certain applications, recognition is feasible with only a few training samples. Secondly, we show that novel objects can be added incrementally without retraining existing objects, which is important for fast interaction. Thirdly, we show that an unbalanced number of positive training samples leads to biased classi er scores that can be corrected by modifying weights. Fourthly, we show that the detector performance can deteriorate due to hard-negative mining for similar or closely related classes (e.g., for Barbie and dress, because the doll is wearing a dress). This can be solved by our hierarchical classi cation. We introduce a new dataset, which we call TOSO, and use it to demonstrate the e ectiveness of the proposed method for the localization and recognition of multiple objects in images.
Appears in Collections:Scholarly Works - FacICTAI

Files in This Item:
File Description SizeFormat 
Incremental_concept_learning_with_few_training_examples _2015.pdf3.97 MBAdobe PDFView/Open

Items in OAR@UM are protected by copyright, with all rights reserved, unless otherwise indicated.