Active Learning on Sparse Graph for Image Annotation = Active Learning on Sparse Graph for Image Annotation

다국어 초록 (Multilingual Abstract)

Due to the semantic gap issue, the performance of automatic image annotation is still far from satisfactory. Active learning approaches provide a possible solution to cope with this problem by selecting most effective samples to ask users to label for training. One of the key research points in active learning is how to select the most effective samples. In this paper, we propose a novel active learning approach based on sparse graph. Comparing with the existing active learning approaches, the proposed method selects the samples based on two criteria: uncertainty and representativeness. The representativeness indicates the contribution of a sample`s label propagating to the other samples, while the existing approaches did not take the representativeness into consideration. Extensive experiments show that bringing the representativeness criterion into the sample selection process can significantly improve the active learning effectiveness.

참고문헌 (Reference)

1 "l1-magic"

2 M. Wang, "Unified video annotation via multi-graph learning" 19 (19): 2009

3 "Trec-10 proceedings appendix on common evaluation measures"

4 N. Roy, "Toward optimal active learning through sampling estimation of error reduction" 441-448, 2001

5 S. Tong, "Support vector machine active learning with applications to text classification" 2 : 45-66, 2002

6 S. Tong, "Support vector machine active learning for image retrieval" 107-118, 2001

7 S. Zhu, "Semi-supervised learning model based efficient image annotation" 16 (16): 989-992, 2009

8 J. Tang, "Semantic-gap-oriented active learning for multilabel image annotation" 21 (21): 2354-2360, 2012

9 Y. Wu, "Sampling strategies for active learning in personal photo retrieval" 529-532, 2006

10 J. Wright, "Robust face recognition via sparse representation" 31 (31): 210-227, 2009

11 Bing-Fei Wu, "Real-Time Vehicle Detector with Dynamic Segmentation and Rule-based Tracking Reasoning for Complex Traffic Conditions" 한국인터넷정보학회 5 (5): 2355-2373, 2011

12 R. Rao, "Probabilistic models of the brain: perception and neural function" MIT Press 2002

13 S. T. Roweis, "Nonlinear dimensionality reduction by locally linear embedding" 290 (290): 2323-2326, 2000

14 T.-S. Chua, "NUS-WIDE: a real-world web image database from National University of Singapore" 2009

15 X. Wang, "Multi-graph enabled active learning for multimodal web image retrieval" 65-72, 2005

16 D. Shen, "Multi-criteria-based active learning for named entity recognition" 589-596, 2004

17 A. Joshi, "Multi-class active learning for image classification" 2372-2379, 2009

18 J. He, "Mean version space: a new active learning method for content-based image retrieval" 15-22, 2004

19 H. Tong, "Manifold-ranking based keyword propagation for image retrieval" 21 : 1-10, 2006

20 C. Dagli, "Leveraging active learning for relevance feedback using an information theoretic diversity measure" 4071 : 123-132, 2006

21 F. Wang, "Label propagation through linear neighborhoods" 20 (20): 55-67, 2008

22 J. Tang, "Inferring semantic concepts from community-contributed images and noisy tags" 223-232, 2009

23 D. Cohn, "Improving generalization with active learning" 15 (15): 201-221, 1994

24 J. Tang, "Image annotation by knn-sparse graph-based label propagation over noisily-tagged web images" 2 (2): 111-126, 2011

25 J. Tang, "Image annotation by graph-based inference with integrated multiple/single instance representations" 12 (12): 131-141, 2010

26 A. Ghoshal, "Hidden Markov models for automatic annotation and content-based retrieval of images and video" 544-551, 2005

27 R. D. King, "Functional genomic hypothesis generation and experimentation by a robot scientist" 427 (427): 247-252, 2004

28 D. L. Donoho, "For most large underdetermined systems of linear equations the minimal -norm solution is also the sparsest solution" 59 (59): 797-829, 2006

29 S. Ayache, "Evaluation of active learning strategies for video indexing" 22 (22): 692-704, 2007

30 M. Wang, "Beyond distance measurement: constructing neighborhood similarity for video annotation" 11 (11): 2009

31 S. C. H. Hoi, "Batch mode active learning and its application to medical image classification" 417-424, 2006

32 C. Zhang, "Annotating retrieval database with active learning" 3 (3): 595-598, 2003

33 X.J. Wang, "Annotating images by mining image search results" 30 (30): 1919-1932, 2008

34 D. Mount, "Ann: a library for approximate nearest neighbor searching" 1997

35 D. A. Cohn, "Active learning with statistical models" 4 (4): 129-145, 1996

36 N. Cesa-Bianchi, "Active learning on trees and graphs" 320-332, 2010

37 M. Wang, "Active learning in multimedia annotation and retrieval: a survey" 2 (2): 2011

38 M. Tang, "Active learning for statistical natural language parsing" 120-127, 2002

39 T. Huang, "Active learning for interactive multimedia retrieval" 96 (96): 648-667, 2008

40 D.D. Lewis, "A sequential algorithm for training text classifiers" 3-12, 1994

41 P. H. Gosselin, "A comparison of active classification methods for content-based image retrieval" 51-58, 2004

42 박재용, "A Novel System for Detecting Adult Images on the Internet" 한국인터넷정보학회 4 (4): 910-924, 2010