TY - GEN
T1 - Multi-class batch-mode active learning for image classification
AU - Joshi, Ajay J.
AU - Porikli, Fatih
AU - Papanikolopoulos, Nikolaos P
N1 - Copyright:
Copyright 2010 Elsevier B.V., All rights reserved.
PY - 2010
Y1 - 2010
N2 - Accurate image classification is crucial in many robotics and surveillance applications - for example, a vision system on a robot needs to accurately recognize the objects seen by its camera. Object recognition systems typically need a large amount of training data for satisfactory performance. The problem is particularly acute when many object categories are present. In this paper we present a batch-mode active learning framework for multi-class image classification systems. In active learning, images are to be chosen for interactive labeling, instead of passively accepting training data. Our framework addresses two important issues: i) it handles redundancy between different images which is crucial when batch-mode selection is performed; and ii) we pose batch-selection as a submodular function optimization problem that makes an inherently intractable problem efficient to solve, while having approximation guarantees. We show results on image classification data in which our approach substantially reduces the amount of training required over the baseline.
AB - Accurate image classification is crucial in many robotics and surveillance applications - for example, a vision system on a robot needs to accurately recognize the objects seen by its camera. Object recognition systems typically need a large amount of training data for satisfactory performance. The problem is particularly acute when many object categories are present. In this paper we present a batch-mode active learning framework for multi-class image classification systems. In active learning, images are to be chosen for interactive labeling, instead of passively accepting training data. Our framework addresses two important issues: i) it handles redundancy between different images which is crucial when batch-mode selection is performed; and ii) we pose batch-selection as a submodular function optimization problem that makes an inherently intractable problem efficient to solve, while having approximation guarantees. We show results on image classification data in which our approach substantially reduces the amount of training required over the baseline.
UR - http://www.scopus.com/inward/record.url?scp=77955800734&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=77955800734&partnerID=8YFLogxK
U2 - 10.1109/ROBOT.2010.5509293
DO - 10.1109/ROBOT.2010.5509293
M3 - Conference contribution
AN - SCOPUS:77955800734
SN - 9781424450381
T3 - Proceedings - IEEE International Conference on Robotics and Automation
SP - 1873
EP - 1878
BT - 2010 IEEE International Conference on Robotics and Automation, ICRA 2010
T2 - 2010 IEEE International Conference on Robotics and Automation, ICRA 2010
Y2 - 3 May 2010 through 7 May 2010
ER -