Abstract
3D biometric techniques on finger traits have become a new trend and have demonstrated a powerful ability for recognition and anti-counterfeiting. Existing methods follow an explicit 3D pipeline that reconstructs the models first and then extracts features from 3D models. However, these explicit 3D methods suffer from the following problems: 1) Inevitable information dropping during 3D reconstruction; 2) Tight coupling between specific hardware and algorithm for 3D reconstruction. It leads us to a question: Is it indispensable to reconstruct 3D information explicitly in recognition tasks? Hence, we consider this problem in an implicit manner, leaving the nerve-wracking 3D reconstruction problem for learnable neural networks with the help of neural radiance fields (NeRFs). We propose FingerNeRF, a novel generalizable NeRF for 3D finger biometrics. To handle the shape-radiance ambiguity problem that may result in incorrect 3D geometry, we aim to involve extra geometric priors based on the correspondence of binary finger traits like fingerprints or finger veins. First, we propose a novel Trait Guided Transformer (TGT) module to enhance the feature correspondence with the guidance of finger traits. Second, we involve extra geometric constraints on the volume rendering loss with the proposed Depth Distillation Loss and Trait Guided Rendering Loss. To evaluate the performance of the proposed method on different modalities, we collect two new datasets: SCUT-Finger-3D with finger images and SCUT-FingerVein-3D with finger vein images. Moreover, we also utilize the UNSW-3D dataset with fingerprint images for evaluation. In experiments, our FingerNeRF can achieve 4.37% EER on SCUT-Finger-3D dataset, 8.12% EER on SCUT-FingerVein-3D dataset, and 2.90% EER on UNSW-3D dataset, showing the superiority of the proposed implicit method in 3D finger biometrics.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Bi, S., Xu, Z., Sunkavalli, K., Hasan, M., Hold-Geoffroy, Y., Kriegman, D., & Ramamoorthi, R. Deep reflectance volumes: Relightable reconstructions from multi-view photometric images. (pp. 294–311).
Chen, A., Xu, Z., Zhao, F., Zhang, X., Xiang, F., Yu, J., & Su, H. (2021). Mvsnerf: Fast generalizable radiance field reconstruction from multi-view stereo. (pp. 14124–14133).
Cheng, K. H. M., Cheng, X., & Zhao, G. (2023). Advancing 3D finger knuckle recognition via deep feature learning. ArXiv arxiv:abs/2301.02934.
Cheng, K. H. M., & Kumar, A. (2020). Contactless biometric identification using 3D finger knuckle patterns. IEEE Transactions on Pattern Analysis and Machine Intelligence, 42, 1868–1883.
Cheng, K. H. M., & Kumar, A. (2021). Deep feature collaboration for challenging 3D finger knuckle identification. IEEE Transactions on Information Forensics and Security, 16, 1158–1173.
Cui, Z., Feng, J., & Zhou, J. (2022). Monocular 3D fingerprint reconstruction and unwarping. IEEE Transactions on Pattern Analysis and Machine Intelligence, 45, 8679–8695.
Deng, K., Liu, A., Zhu, J.-Y., & Ramanan, D. (2022). Depth-supervised nerf: Fewer views and faster training for free. (pp. 12882–12891).
Dong, C., & Kumar, A. M. (2023). Synthesis of multi-view 3D fingerprints to advance contactless fingerprint identification. IEEE Transactions on Pattern Analysis and Machine Intelligence.
Engel, N., Belagiannis, V., & Dietmayer, K. (2021). Point transformer. IEEE Access, 9, 134826–134840.
Guo, M. H., Cai, J. X., Liu, Z. N., Mu, T. J., Martin, R. R., & Hu, S. M. (2021). Pct: Point cloud transformer. Computational Visual Media, 7, 187–199.
He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. (pp. 770–778).
Howard, A., Sandler, M., Chu, G., Chen, L. C., Chen, B., Tan, M., Wang, W., Zhu, Y., Pang, R., Vasudevan, V., & Le, Q.V. (2019). Searching for mobilenetv3. (pp. 1314–1324).
Jain, A. K., Nandakumar, K., & Ross, A. (2016). 50 years of biometric research: Accomplishments, challenges, and opportunities. Pattern Recognition Letters, 79, 80–105.
Kanezaki, A., Matsushita, Y., & Nishida, Y. (2018). Rotationnet: Joint object categorization and pose estimation using multiviews from unsupervised viewpoints. (pp. 5010–5019).
Kang, W., Liu, H., Luo, W., & Deng, F. (2020). Study of a full-view 3D finger vein verification technique. IEEE Transactions on Information Forensics and Security, 15, 1175–1189.
Kızrak, A. M., & Özen, F. (2011). A new median filter based fingerprint recognition algorithm. Procedia Computer Science, 3, 859–865.
Kumar, A., & Kwong, C. (2013). Towards contactless, low-cost and accurate 3D fingerprint identification. (pp. 3438–3443).
Lin, C., & Kumar, A. (2018). Contactless and partial 3D fingerprint recognition using multi-view deep representation. Pattern Recognition, 83, 314–327.
Lin, C., & Kumar, A. (2018). Tetrahedron based fast 3D fingerprint identification using colored leds illumination. IEEE Transactions on Pattern Analysis and Machine Intelligence, 40, 3022–3033.
Lin, C., & Kumar, A. (2018). Matching contactless and contact-based conventional fingerprint images for biometrics identification. IEEE Transactions on Image Processing, 27, 2008–2021.
Liu, J., Ni, B., Li, C., Yang, J., & Tian, Q. (2019). Dynamic points agglomeration for hierarchical point sets learning. (pp. 7546–7555).
Liu, L., Gu, J., Zaw Lin, K., Chua, T.-S., & Theobalt, C. (2020). Neural sparse voxel fields. Advances in Neural Information Processing Systems, 33, 15651–15663.
Lombardi, S., Simon, T., Saragih, J., Schwartz, G., Lehrmann, A., & Sheikh, Y. (2019). Neural volumes: Learning dynamic renderable volumes from images. ACM Transactions on Graphics (TOG), 38, 1–14.
Lowe, D. G. (1999). Object recognition from local scale-invariant features. (Vol. 2, pp. 1150–1157). IEEE.
Ma, X., Qin, C., You, H., Ran, H., & Fu, Y. (2022). Rethinking network design and local geometry in point cloud: A simple residual mlp framework. arXiv preprint arXiv:2202.07123.
Ma, N., Zhang, X., Zheng, H.-T., & Sun, J. (2018). Shufflenet v2: Practical guidelines for efficient cnn architecture design. (pp. 116–131).
Martin-Brualla, R., Radwan, N., Sajjadi, M. S., Barron, J. T., Dosovitskiy, A., & Duckworth, D. (2021). Nerf in the wild: Neural radiance fields for unconstrained photo collections. (pp. 7210–7219).
Mildenhall, B., et al. (2021). Nerf: Representing scenes as neural radiance fields for view synthesis. Communications of the ACM, 65, 99–106.
Miura, N., Nagasaka, A., & Miyatake, T. (2004). Feature extraction of finger-vein patterns based on repeated line tracking and its application to personal identification. Machine Vision and Applications, 15, 194–203.
Qi, C. R., Su, H., Mo, K., & Guibas, L. J. (2017). Pointnet: Deep learning on point sets for 3d classification and segmentation. (pp. 652–660).
Qi, C. R., Yi, L., Su, H., & Guibas, L. J. (2017). Pointnet++: Deep hierarchical feature learning on point sets in a metric space. Advances in Neural Information Processing Systems30.
Qiu, S., Anwar, S., & Barnes, N. (2021). Geometric back-projection network for point cloud classification. IEEE Transactions on Multimedia, 24, 1943–1955.
Radosavovic, I., Kosaraju, R. P., Girshick, R., He, K., & Dollár, P. (2020). Designing network design spaces. (pp. 10428–10436).
Ranftl, R., Lasinger, K., Hafner, D., Schindler, K., & Koltun, V. (2020). Towards robust monocular depth estimation: Mixing datasets for zero-shot cross-dataset transfer. IEEE Transactions on Pattern Analysis and Machine Intelligence, 44, 1623–1637.
Rublee, E., Rabaud, V., Konolige, K., & Bradski, G. (2011). Orb: An efficient alternative to sift or surf. (pp. 2564–2571). IEEE
Sandler, M., Howard, A., Zhu, M., Zhmoginov, A., & Chen, L.-C. (2018). Mobilenetv2: Inverted residuals and linear bottlenecks. (pp. 4510–4520).
Schönberger, J. L. & Frahm, J.-M. (2016). Structure-from-motion revisited .
Schönberger, J. L., Zheng, E., Pollefeys, M., & Frahm, J.-M. (2016). Pixelwise view selection for unstructured multi-view stereo.
Su, H., Maji, S., Kalogerakis, E., & Learned-Miller, E. (2015). Multi-view convolutional neural networks for 3D shape recognition. (pp. 945–953).
Tan, M., & Le, Q. (2019). Efficientnet: Rethinking model scaling for convolutional neural networks. (pp. 6105–6114). (PMLR).
Tan, M., Chen, B., Pang, R., Vasudevan, V., Sandler, M., Howard, A., & Le, Q. V. (2019). Mnasnet: Platform-aware neural architecture search for mobile. (pp. 2820–2828).
Wang, Q., Wang, Z., Genova, K., Srinivasan, P. P., Zhou, H., Barron, J. T., Martin-Brualla, R., Snavely, N., & Funkhouser, T., (2021). Ibrnet: Learning multi-view image-based rendering. (pp. 4690–4699).
Wang, M., & Deng, W. (2021). Deep face recognition: A survey. Neurocomputing, 429, 215–244.
Wang, Y., Sun, Y., Liu, Z., Sarma, S. E., Bronstein, M. M., & Solomon, J. M. (2019). Dynamic graph cnn for learning on point clouds. ACM Transactions on Graphics (tog), 38(5), 1–12.
Xu, M., Zhang, J., Zhou, Z., Xu, M., Qi, X., & Qiao, Y. (May). Learning geometry-disentangled representation for complementary understanding of 3D object point cloud. (Vol. 35, pp. 3056–3064).
Xu, M., Zhou, Z., & Qiao, Y. (2020). Geometry sharing network for 3d point cloud classification and segmentation. (Vol. 34, pp. 12500–12507).
Xu, H., Zhou, Z., Qiao, Y., Kang, W., & Wu, Q. (2021). Self-supervised multi-view stereo via effective co-segmentation and data-augmentation. (Vol. 35, No. 4, pp. 3030–3038).
Xu, H., Zhou, Z., Wang, Y., Kang, W., Sun, B., Li, H., & Qiao, Y. (2021). Digging into uncertainty in self-supervised multi-view stereo. (pp. 6078–6087).
Xu, H., Yang, W., Wu, Q., & Kang, W. (2022). Endowing rotation invariance for 3D finger shape and vein verification. Frontiers of Computer Science, 16, 1–16.
Yang, W., Chen, Z., Huang, J., Wang, L., & Kang, W. (2021). Lfmb-3dfb: A large-scale finger multi-biometric database and benchmark for 3D finger biometrics. (pp. 1–8). IEEEE.
Yang, W., Chen, Z., Huang, J., Wang, L., & Kang, W. (2021). Lfmb-3dfb: A large-scale finger multi-biometric database and benchmark for 3d finger biometrics. 2021 IEEE International Joint Conference on Biometrics (IJCB) (pp. 1–8).
Yang, W., Chen, Z., Huang, J., & Kang, W. (2022). A novel system and experimental study for 3D finger multibiometrics. IEEE Transactions on Biometrics, Behavior, and Identity Science, 4, 471–485.
Yao, Y., Luo, Z., Li, S., Fang, T., & Quan, L. (2018). Mvsnet: Depth inference for unstructured multi-view stereo. (pp. 767–783).
Yao, Y., Luo, Z., Li, S., Shen, T., Fang, T., & Quan, L. (2019). Recurrent mvsnet for high-resolution multi-view stereo depth inference. (pp. 5525–5534).
Yu, T., Meng, J., & Yuan, J. (2018). Multi-view harmonized bilinear network for 3D object recognition. (pp. 186–194).
Yu, A., Ye, V., Tancik, M., & Kanazawa, A. (2021). pixelnerf: Neural radiance fields from one or few images. (pp. 4578–4587).
Zhan, Y., et al. (2020). 3D finger vein biometric authentication with photoacoustic tomography. Applied Optics, 59(28), 8751–8758.
Zhao, H., Jiang, L., Jia, J., Torr, P. H., & Koltun, V. (2021). Point transformer. (pp. 16259–16268).
Zhou, W., Hu, J., Wang, S., Petersen, I. R., & Bennamoun. (2014). Performance evaluation of large 3D fingerprint databases. Electronics Letters, 50, 1060–1061.
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by Segio Escalera.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix A
Appendix A
1.1 Generalizable Neural Rendering
In the main manuscript (Sect. 4.5), we provide the quantitative results of generalizable neural rendering on the test set of 3 different datasets: SCUT-Finger-3D, SCUT-FingerVein-3D, and UNSW-3D. Here, we further provide the quantitative results on the validation set for a complete evaluation. The complete experimental results are provided in the following Tables 17, 18, 19. The quantitative results on SCUT-Finger-3D, SCUT-FingerVein-3D, and UNSW-3D are respectively provided in Tables 17, 18, 19. From the tables, we can witness significant improvement on the validation PSNR, SSIM, and LPIPS on all datasets. This finding is similar as the discussion of the test PSNR, SSIM, and LPIPS on the test set in the main manuscript.
1.2 Ablation Study on Generalizable Neural Rendering
In the main manuscript (Sect. 4.5.6), we provide the ablation results on the test set of the utilized datasets to evaluate the performance on generalizable neural rendering. Here, we further provide the ablation results on the validation set of the three utilized datasets: SCUT-Finger-3D, SCUT-FingerVein-3D, and UNSW-3D. As shown in Tables 20, 21, 22, we respectively present the ablation experiments of the effectiveness of different components in our FingerNeRF on each of the aforementioned datasets. From the tables, we can find that each components can improve the rendering performance effectively on the validation set as well as the test set.
1.3 Multi-view Recognition
In the main manuscript (Sect. 4.6), we provide the comparison with other state-of-the-art multi-view-based recognition methods under the protocol introduced in Sect. 4.6.1. In Tables 23, 24, 25, we respectively show the comparison results on SCUT-Finger-3D, SCUT-FingerVein-3D, and UNSW-3D datasets. In the main manuscript, we select some baseline to save the page space, while we provide a more complete comparison here. As the tables reflects, our method can effectively improve the multi-view recognition results in a generalizable setting.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Xu, H., Huang, J., Ma, Y. et al. Improving 3D Finger Traits Recognition via Generalizable Neural Rendering. Int J Comput Vis 133, 1964–1998 (2025). https://doi.org/10.1007/s11263-024-02248-8
Received:
Accepted:
Published:
Version of record:
Issue date:
DOI: https://doi.org/10.1007/s11263-024-02248-8