Multi-level Adaptive Active Learning for Scene Classification
详细信息 查看全文
- 作者：Xin Li (19)
Yuhong Guo (19) - 关键词：Active Learning ; Scene Classification
- 刊名：Lecture Notes in Computer Science
- 出版年：2014
- 出版时间：2014
- 年：2014
- 卷：8695
- 期：1
- 页码：234-249
- 全文大小：725 KB
- 参考文献：1. Biswas, A., Parikh, D.: Simultaneous active learning of classifiers & attributes via relative feedback. In: Proceedings of CVPR (2013)
2. Dalal, N., Triggs, B.: Histograms of oriented gradients for human detection. In: Proceedings of CVPR (2005)
3. Farabet, C., Couprie, C., Najman, L., LeCun, Y.: Scene parsing with multiscale feature learning, purity trees,and optimal covers. CoRR abs/1202.2160 (2012)
4. Fei-Fei, P.L., Perona: A bayesian hierarchical model for learning natural scene categories. In: Proceedings of CVPR (2005)
5. Gould, S., Gao, T., Koller, D.: Region-based segmentation and object detection. In: Proceedings of NIPS (2009)
6. Guo, Y., Greiner, R.: Optimistic active learning using mutual information. In: Proceedings of IJCAI (2007)
7. Jain, P., Kapoor, A.: Active learning for large multi-class problems. In: Proceedings of CVPR (2009)
8. Joshi, A., Porikli, F., Papanikolopoulos, N.: Multi-class active learning for image classification. In: Proceedings of CVPR (2009)
9. Kapoor, A., Grauman, K., Urtasun, R., Darrell, T.: Active learning with gaussian processes for object categorization. In: Proceedings of ICCV (2007)
10. A., Kapoor, G.H., Akbarzadeh, A., Baker, S.: Which faces to tag: Adding prior constraints into active learning. In: Proceedings of ICCV (2009)
11. Kovashka, A., Vijayanarasimhan, S., Grauman, K.: Actively selecting annotations among objects and attributes. In: Proceedings of ICCV (2011)
12. Kumar, M., Koller, D.: Efficiently selecting regions for scene understanding. In: Proceedings of CVPR (2010)
13. Lazebnik, S., Schmid, C., Ponce, J.: Beyond bags of features: Spatial pyramid matching for recognizing natural scene categories. In: Proceedings of CVPR (2006)
14. Li, L., Su, H., Xing, E., Fei-Fei, L.: Object bank: A high-level image representation for scene classification & semantic feature sparsification. In: Proceedings of NIPS (2010)
15. Li, X., Guo, Y.: Adaptive active learning for image classification. In: Proceedings of CVPR (2013)
16. Lin, C., Weng, R., Keerthi, S.: Trust region newton method for logistic regression. J. Mach. Learn. Res. 9 (June 2008)
17. Lowe, D.: Distinctive image features from scale-invariant keypoints. Int. J. Comput. Vision 60(2) (November 2004)
18. Mensink, T., Verbeek, J., Csurka, G.: Learning structured prediction models for interactive image labeling. In: Proceedings of CVPR (2011)
19. Pandey, M., Lazebnik, S.: Scene recognition and weakly supervised object localization with deformable part-based models. In: Proceedings of ICCV (2011)
20. Parizi, S., Oberlin, J., Felzenszwalb, P.: Reconfigurable models for scene recognition. In: Proceedings of CVPR (2012)
21. Parkash, A., Parikh, D.: Attributes for classifier feedback. In: Fitzgibbon, A., Lazebnik, S., Perona, P., Sato, Y., Schmid, C. (eds.) ECCV 2012, Part III. LNCS, vol.?7574, pp. 354-68. Springer, Heidelberg (2012) CrossRef
22. Patterson, G., Hays, J.: Sun attribute database: Discovering, annotating, and recognizing scene attributes. In: Proceeding of CVPR (2012)
23. Quattoni, A., Torralba, A.: Recognizing indoor scenes. In: Proceedings of CVPR (2009)
24. Sadeghi, F., Tappen, M.F.: Latent pyramidal regions for recognizing scenes. In: Fitzgibbon, A., Lazebnik, S., Perona, P., Sato, Y., Schmid, C. (eds.) ECCV 2012, Part V. LNCS, vol.?7576, pp. 228-41. Springer, Heidelberg (2012) CrossRef
25. Settles, B.: Active Learning. Synthesis digital library of engineering and computer science. Morgan & Claypool (2011)
26. Sharmanska, V., Quadrianto, N., Lampert, C.H.: Augmented attribute representations. In: Fitzgibbon, A., Lazebnik, S., Perona, P., Sato, Y., Schmid, C. (eds.) ECCV 2012, Part V. LNCS, vol.?7576, pp. 242-55. Springer, Heidelberg (2012) CrossRef
27. Siddiquie, B., Gupta, A.: Beyond active noun tagging: Modeling contextual interactions for multi-class active learning. In: Proceedings of CVPR (2010)
28. Jones, K.S., Willett, P.: Readings in Information Retrieval. Morgan Kaufmann Publishers Inc. (1997)
29. Vezhnevets, A., Buhmann, J., Ferrari, V.: Active learning for semantic segmentation with expected change. In: Proceedings of CVPR (2012)
30. Vijayanarasimhan, S., Grauman, K.: Multi-level active prediction of useful image annotations for recognition. In: Proceedings of NIPS (2008)
31. Vijayanarasimhan, S., Grauman, K.: Large-scale live active learning: Training object detectors with crawled data and crowds. In: Proceedings of CVPR (2011)
32. Wang, Y., Mori, G.: A discriminative latent model of object classes and attributes. In: Daniilidis, K., Maragos, P., Paragios, N. (eds.) ECCV 2010, Part V. LNCS, vol.?6315, pp. 155-68. Springer, Heidelberg (2010) CrossRef
33. Wu, J., Rehg, J.: CENTRIST: A Visual Descriptor for Scene Categorization. IEEE Transactions on PAMI 33 (2011)
34. Yan, A.R., Yang, L., Hauptmann: Automatically labeling video data using multi-class active learning. In: Proceedings of ICCV (2003) - 作者单位：Xin Li (19)
Yuhong Guo (19)
19. Department of Computer and Information Sciences, Temple University, Philadelphia, PA, 19122, USA - ISSN：1611-3349
文摘
Semantic scene classification is a challenging problem in computer vision. In this paper, we present a novel multi-level active learning approach to reduce the human annotation effort for training robust scene classification models. Different from most existing active learning methods that can only query labels for selected instances at the target categorization level, i.e., the scene class level, our approach establishes a semantic framework that predicts scene labels based on a latent object-based semantic representation of images, and is capable to query labels at two different levels, the target scene class level (abstractive high level) and the latent object class level (semantic middle level). Specifically, we develop an adaptive active learning strategy to perform multi-level label query, which maintains the default label query at the target scene class level, but switches to the latent object class level whenever an “unexpected-target class label is returned by the labeler. We conduct experiments on two standard scene classification datasets to investigate the efficacy of the proposed approach. Our empirical results show the proposed adaptive multi-level active learning approach can outperform both baseline active learning methods and a state-of-the-art multi-level active learning method.