Abstract
Target recognition difficulty quantification and prediction using the search time for the human visual system to target an object is a challenging task, which can effectively guide the training of machine learning models such as target recognition and target location. Our work focuses on how to use region-of-interest (ROI) information to improve the accuracy of the visual search difficulty prediction model. First, the influence of ROI information on visual search difficulty is explored in this paper. Then, based on the learning using privileged information paradigm, we build a support vector regression model using privileged information (SVR +), which uses the deep features of ROIs in the training stage. Next, a coordinate descent algorithm is developed to solve the dual optimization problem in SVR + training. Comprehensive experiments validate the improvement in the accuracy of the proposed model in predicting the difficulty of visual search and the efficiency of our coordinate descent algorithm in model training.
Similar content being viewed by others
References
Ionescu RT, Alexe B, Leordeanu M, Popescu M, Papadopoulos D, Ferrari V (2016) How hard can it be? Estimating the difficulty of visual search in an image [C]. IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 1:2157–2166
Vapnik V, Vashist A (2009) A new learning paradigm: learning using privileged information [J]. Neural Netw 22(5–6):544–557
Tian Y, Mirzabagheri M, Tirandazi P, Bamakan SMH (2020) A non-convex semi-supervised approach to opinion spam detection by ramp-one class SVM. Inf Process Manag 57(6):102381
Li W, Niu L, Xu D (2014) Exploiting Privileged Information from Web Data for Image Categorization [C]. IEEE Conference on European Conference on Computer Vision (ECCV). https://doi.org/10.1007/978-3-319-10602-1_29
Şahin DÖ, Kural OE, Akleylek S et al (2021) A novel permission-based Android malware detection system using feature selection based on linear regression. Neural Comput Applic. https://doi.org/10.1007/s00521-021-05875-1
Sadjadi FA, Bazakos M (1991) A perspective on automatic target recognition evaluation technology [J]. Opt Eng 30(2):2–15
Peters RA II, Strickland RN (1990) Image complexity metrics for automatic target recognizers [C]. ATR System and Technology Conference 1:1–17
Mario I, Chacon M, Alma D et al (2005) Image complexity measure: a human criterion free approach [C]. IEEE Nafips 2005 Meeting North American. 1:241–246
Vijayanarasimhan S, Grauman K (2009) What's it going to cost you? Predicting effort vs. informativeness for multi-label image annotations [C]. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2262–2269.
Feyereisl J, Aickelin U (2012) Privileged information for data clustering [J]. Inf Sci 194(5):4–23
Ji Y, Sun S, Lu Y. (2013) Multitask multiclass privileged information support vector machines[C]. Proceedings of the 21st International Conference on Pattern Recognition (ICPR2012). IEEE
Wang S, Menghua HE, Zhu Y et al (2015) Learning with privileged information using Bayesian networks [J]. Front Comp Sci 9(002):185–199
Xu X, Li W, Xu D (2015) Distance metric learning using privileged information for face verification and person re-identification [J]. IEEE Transactions on Neural Networks & Learning Systems 26(12):3150
Pechyony D, Izmailov R, Vashist A, et al. (2011) SMO-Style Algorithms for Learning Using Privileged Information [C]. International Conference on Data Mining, 235–241.
Pechyony D, Vapnik V. (2011) Fast Optimization Algorithms for Solving SVM+ [J]. Statiscal Learning
Aulia Khilmi Rizgi, Anhar Risnumawan, Fernando Ardila, Edi Sutoyo, Ryan Satria Wijaya, Ilham Fakhrul Arifin, Martianda Erste Anggraeni, Tutut Herawan. (2020) Visual perception system of EROS humanoid robot soccer. Int. J. Intell. Inf. Technol. 16(4): 68–86
Bouwmans T, Silva C, Marghes C et al (2016) On the role and the importance of features for background modeling and foreground detection [J]. Computer Science Review 28:26–91
Wright SJ (2015) Coordinate descent algorithms [J]. Math Program 151(1):3–34
Fizza K, Banerjee A, Mitra K et al (2021) QoE in IoT: a vision, survey and future directions. Discov Internet Things 1:4
K. Simonyan, A. Zisserman. (2014) Very deep convolutional networks for large-scale image recognition [J], Computer Science
K. He, X. Zhang, S. Ren, J. Sun. (2016) Deep residual learning for image recognition [C]. IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
Chang C-C, Lin C-J (2002) Training ν-support vector regression: Theory and algorithms [J]. Neural Comput 14(8):1959–1977
Yantao R, Yuchang H, Xue S (2006) Saccade and its mechanism in visual search [J]. Advance in Psychological Science 14(03):340–345
Gilchrist ID, Harvey M (2000) Refixation frequency and memory mechanisms in visual search [J]. Curr Biol 10(19):1209–1212
Acknowledgements
This research was funded by China Postdoctoral Science Foundation, Grant Number 2020M673606XB
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest
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
Xiao, B., Liu, X. & Wang, C. Visual search difficulty prediction with image ROI information. Neural Comput & Applic 34, 6799–6809 (2022). https://doi.org/10.1007/s00521-021-06413-9
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00521-021-06413-9