Abstract
The huge data resource on the Web provides us with an emerging chance to solve the lack of training sample problem that lasting for years in facial age estimation. We show that the web images assisted age estimation can be modeled into a Multiple Instance Regression (MIR) problem. Different from the traditional Multiple Instance Learning (MIL) problem that deals with the bag-level classification task, we model the age estimation problem as an instance-level task. To this end, it is essential to reveal the latent instance labels from all bags. In this paper, we propose a novel algorithm named Witness Detecting Multi-instance Regression (WDMR) that can find all possible positive instances from training bags and use them to train an instance-level regressor. Considering the connection between neighbor relationship and inter-/intra- class differences among instances, we develop a Supervised Citer k-Nearest Neighbor (SC-kNN) graph and a sparse voting strategy to address these problems within a joint learning framework. Experimental results on synthetic and real-world datasets have verified the advantages of our method compared with other state-of-the-art MIL approaches.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Andrews S, Tsochantaridis I, Hofmann T (2003) Support vector machines for multiple-instance learning. In: Annual conference on neural information processing systems (NIPS), pp 1–8
Astorino A, Fuduli A, Gaudioso M (2019) A lagrangian relaxation approach for binary multiple instance classification. IEEE Trans Neural Netw Learning Syst (99):1–10
Babenko B, Yang MH, Belongie S (2011) Robust object tracking with online multiple instance learning. IEEE Trans Pattern Anal Mach Intell 33(8):1619–1632
Dietterich T, Lathrop R, Lozano-Prez T (1997) Solving the multiple-instance problem with axis-parallel rectangles. Artif Intell:31–71
Fu Y, Guo G, Huang TS (2010) Age synthesis and estimation via faces: a survey. IEEE Trans Pattern Anal Mach Intell (t-PAMI) 32(11):1955–1976
Geng X, Yin C, Zhou Z (2013) Facial age estimation by learning from label distributions. IEEE Trans Pattern Anal Mach Intell (t-PAMI) 35(10):2401–2412
Guillaumin M, Verbeek J, Schmid C (2009) Is that you? metric learning approaches for face identification. IEEE Int Conf Comput Vision (ICCV):1–8
He Z, Li X, Zhang Z, Wu F, Geng X, Zhang Y, Yang MH, Zhuang Y (2017) Data-dependent label distribution learning for age estimation. IEEE Trans Image Process 26(8):3846–3858
Jia M, Gao Z, Guo Q, Gu X (2017) Fast compressive tracking with robust example selection based on multiple instance learning in smart and autonomous systems. Pattern Recognition
Kraus OZ, Ba JL, Frey BJ (2016) Classifying and segmenting microscopy images with deep multiple instance learning. Bioinformatics 32(12):i52–i59
Lan X, Ye M, Shao R, Zhong B, Yuen PC, Zhou H (2019) Learning modality-consistency feature templates: a robust rgb-infrared tracking system. IEEE Trans Ind Electron:1–1
Lan X, Yuen PC, Chellappa R (2017) Robust mil-based feature template learning for object tracking. In: The 31st AAAI conference on artificial intelligence, pp 4118–4125
Lan X, Zhang S, Yuen PC, Chellappa R (2018) Learning common and feature-specific patterns: a novel multiple-sparse-representation-based tracker. IEEE Trans Image Process 27(4):2022–2037
Li K, Xing J, Su C, Hu W, Zhang Y, Maybank S (2018) Deep cost-sensitive and order-preserving feature learning for cross-population age estimation. In: IEEE conference on computer vision and pattern recognition (CVPR), pp 399–408
Li YF, Kwok J, Tsang I, Zhou Z (2009) A convex method for locating regions of interest with multi-instance learning. European Conference on Machine Learning (ECML):15–30
Liao H (2019) Facial age feature extraction based on deep sparse representation. Multimed Tools Appl 78(2):2181–2197
Liu G, Wu J, Zhou Z (2012) Key instance detection in multi-instance learning. Asian Conf Machine Learning (ACML):1–16
Ma Y, Liu J, Yang X, Liu Y, Zheng N (2015) Double layer multiple task learning for age estimation with insufficient training samples. Neurocomputing 147:380–386
Mathias M, Benenson R, Pedersoli M, Gool LV (2014) Face detection without bells and whistles. In: European conference on computer vision
Ni B, Song Z, Yan S (2011) Web image and video mining towards universal and robust age estimator. IEEE Trans Multimed (t-MM) 13(6):1217–1229
Parkhi OM, Vedaldi A, Zisserman A (2015) Deep face recognition. In: British machine vision conference
Wang D, Li J, Zhang B (2006) Multiple-instance learning via random walk. European Conference on Machine Learning (ECML):473–484
Wang J, Zucker JD (2000) Solving the multiple-instance problem: a lazy learning approach. International Conference on Machine Learning (ICML):1–7
Yang X, Liu J, Ma Y, Xue J (2014) Facial age estimation from web photos using multiple-instance learning. IEEE Int Conf Multimed Expo (ICME):1–6
Zhang K, Song H (2013) Real-time visual tracking via online weighted multiple instance learning. Pattern Recogn 46(1):397–411
Zhong P, Gong Z, Shan J (2019) Multiple instance learning for multiple diverse hyperspectral target characterizations. IEEE Trans Neural Netw Learning Syst (99):1–13
Acknowledgements
This work is supported by the funding of China Scholarship Council (CSC) (No.201706285003), the Provincial Key Laboratory Program of Shaanxi (No. 2013SZS12-K01), and the National Natural Scientific Foundation of China (No.61379104).
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Liu, J., Qiao, R., Li, Y. et al. Witness detection in multi-instance regression and its application for age estimation. Multimed Tools Appl 78, 33703–33722 (2019). https://doi.org/10.1007/s11042-019-08203-x
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-019-08203-x