Abstract
The use of deep learning in computer vision has been extremely successful. Nevertheless, for the tattoo recognition task very few approaches incorporate this technique, mainly due to the lack of large datasets to train specific models. Within this domain, some works have used the intermediate layers of pre-trained object classification networks to extract a global tattoo image descriptor, avoiding the expensive work of training from scratch. Although that approach showed good results, it does not incorporate specific knowledge of tattoo identification. In this work, we propose an attention pooling method that addresses this problem. Our method uses several functions to weight the local features of a convolutional feature map and then those weights are averaged using again some weights associated with each function. The use of these weighting functions provides more or less importance to local regions of the tattoo image, allowing the recognition process to take into account some domain-specific characteristics. This approach showed promising results in three tattoo databases, outperforming previous state-of-the-art works.
Similar content being viewed by others
References
Babenko A, Lempitsky V (2015) Aggregating local deep features for image retrieval. In: Proceedings of the IEEE international conference on computer vision, pp 1269–1277
Berno B C S, et al. (2021) Sketch-based multimodal image retrieval using deep learning. Master’s Thesis, Universidade Tecnológica Federal do Paraná
Dantcheva A, Elia P, Ross A (2016) What else does your biometric data reveal? a survey on soft biometrics. IEEE Trans Inf Forensic Secur 11 (3):441–467
Di X, Patel V M (2017) Deep learning for tattoo recognition. In: Deep Learning for biometrics. Springer, Cham, pp 241–256
Han H, Li J, Jain A K, Chen X (2019) Tattoo image search at scale: Joint detection and compact representation learning. IEEE Transactions on Pattern Analysis and Machine Intelligence
Horta A A, Magalhães L P (2018) Comparing the performance and accuracy of algorithms applied to tattoos images identification
Hrkac T, Brkic K, Ribaric S, Marcetic D (2016) Deep learning architectures for tattoo detection and de-identification. In: SPLINE, pp 1–5
Jain A K, Lee J-E, Jin R (2007) Tattoo-id: Automatic tattoo image retrieval for suspect and victim identification. In: Pacific-Rim Conference on Multimedia, pp 256–265
Jiawang C, Yuan Z (2018) Tattoo recognition based on triplet gan. In: 2018 37th Chinese Control Conference (CCC). IEEE, pp 9595–9597
Jimenez A, Alvarez J M, Giro-i Nieto X (2017) Class-weighted convolutional features for visual instance search. arXiv:1707.02581
Kalantidis Y, Mellina C, Osindero S (2016) Cross-dimensional weighting for aggregated deep convolutional features. In: European conference on computer vision, pp 685–701
Kim J, Parra A, Yue J, Li H, Delp E J (2015) Robust local and global shape context for tattoo image matching. In: Image Processing (ICIP), 2015 IEEE International Conference on, pp 2194–2198
Kolkur S, Kalbande D, Shimpi P, Bapat C, Jatakia J (2017) Human skin detection using rgb, hsv and ycbcr color models. arXiv:1708.02694
Lee J-E, Jain A K, Jin R (2008) Scars, marks and tattoos (smt): Soft biometric for suspect and victim identification. In: Biometrics Symposium, 2008. BSYM’08, pp 1–8
Lowe D G (2004) Distinctive image features from scale-invariant keypoints. Int J Comput Vis 60(2):91–110
Manger D (2012) Large-scale tattoo image retrieval. In: Computer and Robot Vision (CRV), 2012 Ninth Conference on, pp 454–459
Nicolás-Díaz M, Morales-González A, Méndez-Vázquez H (2019) Deep generic features for tattoo identification. In: Iberoamerican Congress on Pattern Recognition, pp 272–282
Sandler M, Howard A, Zhu M, Zhmoginov A, Chen L-C (2018) Mobilenetv2: Inverted residuals and linear bottlenecks. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 4510–4520
Sun Z H, Baumes J, Tunison P, Turek M, Hoogs A (2016) Tattoo detection and localization using region-based deep learning. In: Pattern Recognition (ICPR), 2016 23rd International Conference on, pp 3055–3060
Wan W, Chen J, Li T, Huang Y, Tian J, Yu C, Xue Y (2019) Information entropy based feature pooling for convolutional neural networks. In: Proceedings of the IEEE international conference on computer vision, pp 3405–3414
Wu X, Irie G, Hiramatsu K, Kashino K (2018) Weighted generalized mean pooling for deep image retrieval. In: 2018 25th IEEE International Conference on Image Processing (ICIP), pp 495–499
Xu Q, Ghosh S, Xu X, Huang Y, Kong A W K (2016) Tattoo detection based on cnn and remarks on the nist database. In: Biometrics (ICB), 2016 International Conference on, pp 1–7
Xu X, Kong A W K (2016) A geometric-based tattoo retrieval system. In: Pattern Recognition (ICPR), 2016 23rd International Conference on, pp 3019–3024
Zhou B, Khosla A, Lapedriza A, Oliva A, Torralba A (2016) Learning deep features for discriminative localization. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 2921–2929
Author information
Authors and Affiliations
Corresponding authors
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
Nicolás-Díaz, M., Morales-González, A. & Méndez-Vázquez, H. Weighted average pooling of deep features for tattoo identification. Multimed Tools Appl 81, 25853–25875 (2022). https://doi.org/10.1007/s11042-022-12516-9
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-022-12516-9