Abstract
Hierarchical classification (HC) is an effective method to solve the problem of multi-class classification especially when the categories are organized hierarchically. However, HC models perform worse than flat classification (FC) models due to blocking, i.e. errors occur at the higher level will be propagated to the lower level. The first step in solving the blocking problem is to capture the blocked samples. In this paper, we present a novel HC model to capture the blocked samples by adding a virtual dustbin category for each middle-layer classifier, referred to as DBHC model. Furthermore, in order to improve the classification accuracy and accelerate the convergence process, we propose a feedback incremental learning (FIL) strategy to take into account the weights of samples, which can adjust the composition of training samples according to the test results of the previous training steps. Experiments on fashion image classification shows the superiority of the proposed model compared with several prior methods.
The first author is a student.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Krizhevsky, A., Sutskever, I., Hinton, G.E.: Imagenet classification with deep convolutional neural networks. In: Advances in Neural Information Processing Systems, pp. 1097–1105 (2012)
Yang, X., et al.: Co-trained convolutional neural networks for automated detection of prostate cancer in multi-parametric MRI. Med. Image Anal. 42, 212–227 (2017)
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 770–778 (2016)
Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556 (2014)
Moeslund, T.B., Granum, E.: A survey of computer vision-based human motion capture. Comput. Vis. Image Underst. 81(3), 231–268 (2001)
Bhaskar, H., Hoyle, D.C., Singh, S.: Machine learning in bioinformatics: a brief survey and recommendations for practitioners. Comput. Biol. Med. 36(10), 1104–1125 (2006)
Greiner, R., Grove, A., Schuurmans, D.: On learning hierarchical classifications. In: ResearchIndex; The NECI Scientifc Literature Digital Libraray, Citado em, vol. 32, pp. 34–40 (1997)
Dumais, S., Chen, H.: Hierarchical classification of web content. In: Proceedings of the 23rd Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, pp. 256–263. ACM (2000)
Sun, A., Lim, E.P., Ng, W.K., Srivastava, J.: Blocking reduction strategies in hierarchical text classification. IEEE Trans. Knowl. Data Eng. 16(10), 1305–1308 (2004)
Wen, L., Duo-Qian, M., Zhi-Hua, W., Wei-Li, W.: Hierarchical text classification model based on blocking priori knowledge. Pattern Recognit. Artif Intell. 23(4), 456–463 (2010)
Koller, D., Sahami, M.: Hierarchically classifying documents using very few words. Technical report, Stanford InfoLab (1997)
Liu, T.Y., Yang, Y., Wan, H., Zeng, H.J., Chen, Z., Ma, W.Y.: Support vector machines classification with a very large-scale taxonomy. ACM SIGKDD Explora. Newsl. 7(1), 36–43 (2005)
Xue, G.R., Xing, D., Yang, Q., Yu, Y.: Deep classification in large-scale text hierarchies. In: Proceedings of the 31st Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, pp. 619–626. ACM (2008)
Zhu, S., Wei, X.Y., Ngo, C.W.: Collaborative error reduction for hierarchical classification. Comput. Vis. Image Underst. 124, 79–90 (2014)
Naik, A., Rangwala, H.: Improving large-scale hierarchical classification by rewiring: a data-driven filter based approach. J. Intell. Inf. Syst. 52(1), 141–164 (2019)
Naik, A., Rangwala, H.: Filter based taxonomy modification for improving hierarchical classification. arXiv preprint arXiv:1603.00772 (2016)
Nooka, S.P., Chennupati, S., Veerabhadra, K., Sah, S., Ptucha, R.: Adaptive hierarchical classification networks. In: 2016 23rd International Conference on Pattern Recognition (ICPR), pp. 3578–3583. IEEE (2016)
Wang, X., Zhang, T.: Clothes search in consumer photos via color matching and attribute learning. In: Proceedings of the 19th ACM international conference on Multimedia, pp. 1353–1356. ACM (2011)
Fu, J., Wang, J., Li, Z., Xu, M., Lu, H.: Efficient clothing retrieval with semantic-preserving visual phrases. In: Lee, K.M., Matsushita, Y., Rehg, J.M., Hu, Z. (eds.) ACCV 2012. LNCS, vol. 7725, pp. 420–431. Springer, Heidelberg (2013). https://doi.org/10.1007/978-3-642-37444-9_33
Liu, Z., Luo, P., Qiu, S., Wang, X., Tang, X.: DeepFashion: powering robust clothes recognition and retrieval with rich annotations. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 1096–1104 (2016)
Di, W., Wah, C., Bhardwaj, A., Piramuthu, R., Sundaresan, N.: Style finder: fine-grained clothing style detection and retrieval. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, pp. 8–13 (2013)
Ke, S.W., Lin, W.C., Tsai, C.F., Hu, Y.H.: Soft estimation by hierarchical classification and regression. Neurocomputing 234, 27–37 (2017)
Naik, A., Rangwala, H.: Inconsistent node flattening for improving top-down hierarchical classification. In: 2016 IEEE International Conference on Data Science and Advanced Analytics (DSAA), pp. 379–388. IEEE (2016)
Acknowledgment
The work is partially supported by the National Key Research and Development Project (No. 213), the National Nature Science Foundation of China (No. 61573259, No. 61673299, No. 61673301, No. 61573255), the Key Lab of Information Network Security, Ministry of Public Security (No. C18608) and Shanghai Health and Family Planning Commission Chinese Medicine and Technology Innovation Project (No. ZYKC201702005).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this paper
Cite this paper
Chen, Y., Shen, W., Li, Q., Wei, Z. (2019). A Dustbin Category Based Feedback Incremental Learning Strategy for Hierarchical Image Classification. In: Lin, Z., et al. Pattern Recognition and Computer Vision. PRCV 2019. Lecture Notes in Computer Science(), vol 11857. Springer, Cham. https://doi.org/10.1007/978-3-030-31654-9_41
Download citation
DOI: https://doi.org/10.1007/978-3-030-31654-9_41
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-31653-2
Online ISBN: 978-3-030-31654-9
eBook Packages: Computer ScienceComputer Science (R0)