Abstract
Annotating age classes for humans’ facial images according to their appearance is very challenging because of dynamic person-specific ageing pattern, and thus leads to a set of unreliable apparent age labels for each image. For utilising ambiguous label annotations, an intuitive strategy is to generate a pseudo age for each image, typically the average value of manually-annotated age annotations, which is thus fed into standard supervised learning frameworks designed for chronological age estimation. Alternatively, inspired by the recent success of label distribution learning, this paper introduces a novel concept of ambiguous label distribution for apparent age estimation, which is developed under the following observations that (1) soft labelling is beneficial for alleviating the suffering of inaccurate annotations and (2) more reliable annotations should contribute more. To achieve the goal, label distributions of sparse age annotations for each image are weighted according to their reliableness and then combined to construct an ambiguous label distribution. In the light, the proposed learning framework not only inherits the advantages from conventional learning with label distribution to capture latent label correlation but also exploits annotation reliableness to improve the robustness against inconsistent age annotations. Experimental evaluation on the FG-NET age estimation benchmark verifies its effectiveness and superior performance over the state-of-the-art frameworks for apparent age estimation.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Following the usage in Matlab, the notation [x : y : z] represents an array starting from x to z with the step of y.
References
Chen, K., Gong, S., Xiang, T., Loy, C.C.: Cumulative attribute space for age and crowd density estimation. In: CVPR (2013)
Fu, Y., Guo, G., Huang, T.S.: Age synthesis and estimation via faces: a survey. TPAMI 32(11), 1955–1976 (2010)
Geng, X., Ji, R.: Label distribution learning. In: ICDMW (2013)
Geng, X., Yin, C., Zhou, Z.H.: Facial age estimation by learning from label distributions. TPAMI 35(10), 2401–2412 (2014)
Luu, K., Ricanek Jr., K., Bui, T.D., Suen, C.Y.: Age estimation using active appearance models and support vector machine regression. In: BTAS (2009)
Pontes, J.K., Britto, A.S., Fookes, C., Koerich, A.L.: A flexible hierarchical approach for facial age estimation based on multiple features. Pattern Recogn. 54, 34–51 (2015)
Zhang, Y., Yeung, D.: Multi-tasks warped Gaussian process for personalized age estimation. In: CVPR (2010)
Han, H., Otto, C., Liu, X., Jain, A.K.: Demographic estimation from face images: human vs. machine performance. TPAMI 37(6), 1148–1161 (2015)
Yang, X., Gao, B.B., Xing, C., Huo, Z.W., Wei, X.S., Zhou, Y., Wu, J., Geng, X.: Deep label distribution learning for apparent age estimation. In: CVPR Workshops (2015)
Rothe, R., Timofte, R., Gool, L.: DEX: deep expectation of apparent age from a single image. In: ICCV Workshops (2015)
Kuang, Z., Huang, C., Zhang, W.: Deeply learned rich coding for cross-dataset facial age estimation. In: ICCV Workshops (2015)
Zhu, Y., Li, Y., Mu, G., Guo, G.: A study on apparent age estimation. In: ICCV Workshops (2015)
Antipov, G., Baccouche, M., Berrani, S.A., Dugelay, J.L.: Apparent age estimation from face images combining general and children-specialized deep learning models. In: ICCV Workshops (2015)
Geng, X., Zhou, Z.H., Smith-Miles, K.: Automatic age estimation based on facial aging patterns. TPAMI 29(12), 2234–2240 (2007)
Lanitis, A., Draganova, C., Christodoulou, C.: Comparing different classifiers for automatic age estimation. TSMC 34(1), 621–628 (2004)
Geng, X., Wang, Q., Xia, Y.: Facial age estimation by adaptive label distribution learning. In: ICPR (2014)
Guo, G., Fu, Y., Huang, T.S., Dyer, C.R.: Image-based human age estimation by manifold learning and locally adjusted robust regression. TIP 17(7), 1178–1188 (2008)
Guo, G., Mu, G., Fu, Y., Huang, T.S.: Human age estimation using bio-inspired features. In: CVPR (2009)
Chang, K.Y., Chen, C.S., Hung, Y.P.: Ordinal hyperplanes ranker with cost sensitivities for age estimation. In: CVPR (2011)
Wang, S., Tao, D., Yang, J.: Relative attribute SVM+ learning for age estimation. TC 46(3), 827–839 (2015)
Zhang, M.L., Yu, F.: Solving the partial label learning problem: an instance-based approach. In: IJCAI (2015)
Cour, T., Sapp, B., Taskar, B.: Learning from partial labels. JMLR 12, 1501–1536 (2011)
Cour, T., Sapp, B., Jordan, C., Taskar, B.: Learning from ambiguously labeled images. In: CVPR (2009)
Escalera, S., Fabian, J., Pardo, P., Baro, X., Gonzalez, J., Escalante, H., Guyon, I.: ChaLearn 2015 apparent age and cultural event recognition: datasets and results. In: ICCV, ChaLearn Looking at People Workshop (2015)
Krizhevsky, A., Sutskever, I., Hinton, G.E.: Imagenet classification with deep convolutional neural networks. In: NIPS (2012)
MacKay, D.J.: Information Theory, Inference and Learning Algorithms. Cambridge University Press, Cambridge (2003)
Berger, A.L., Pietra, V.J.D., Pietra, S.A.D.: A maximum entropy approach to natural language processing. Comput. Linguist. 22(1), 39–71 (1996)
Pietra, S.D., Pietra, V.D., Lafferty, J.: Inducing features of random fields. TPAMI 19(4), 380–393 (1997)
Nocedal, J., Wright, S.: Numerical Optimization. Springer Science & Business Media, New York (2006)
Malouf, R.: A comparison of algorithms for maximum entropy parameter estimation. In: The 6th Conference on Natural Language Learning (2002)
Cootes, T.F., Edwards, G.J., Taylor, C.J.: Active appearance models. TPAMI 23(6), 681–685 (2001)
Yan, S., Wang, H., Huang, T.S., Yang, Q., Tang, X.: Ranking with uncertain labels. In: ICME (2007)
Yan, S., Wang, H., Tang, X., Huang, T.S.: Learning auto-structured regressor from uncertain nonnegative labels. In: ICCV (2007)
Smola, A.J., Schölkopf, B.: A tutorial on support vector regression. Stat. Comput. 14(3), 199–222 (2004)
Acknowledgement
This work was funded by Academy of Finland under the Grant No. 267581 and 298700, and D2I SHOK project funded by Digile Oy and Nokia Technologies (Tampere, Finland). The authors wish to acknowledge CSC-IT Center for Science, Finland, for generous computational resources.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer International Publishing AG
About this paper
Cite this paper
Chen, K., Kämäräinen, JK. (2017). Learning with Ambiguous Label Distribution for Apparent Age Estimation. In: Lai, SH., Lepetit, V., Nishino, K., Sato, Y. (eds) Computer Vision – ACCV 2016. ACCV 2016. Lecture Notes in Computer Science(), vol 10113. Springer, Cham. https://doi.org/10.1007/978-3-319-54187-7_22
Download citation
DOI: https://doi.org/10.1007/978-3-319-54187-7_22
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-54186-0
Online ISBN: 978-3-319-54187-7
eBook Packages: Computer ScienceComputer Science (R0)