Bayesian Feature Fusion Using Factor Graph in Reduced Normal Form
Source abstract
Abstract
In this work, we investigate an Information Fusion architecture based on a Factor Graph in Reduced Normal Form. This paradigm permits to describe the fusion in a completely probabilistic framework and the information related to the different features are represented as messages that flow in a probabilistic network. In this way we build a sort of context for observed features conferring to the solution a great flexibility for managing different type of features with wrong and missing values as required by many real applications. Moreover, modifying opportunely the messages that flow into the network, we obtain an effective way to condition the inference based on the different reliability of each information source or in presence of single unreliable signal. The proposed architecture has been used to fuse different detectors for an identity document classification task but its flexibility, extendibility and robustness make it suitable to many real scenarios where the signal can be wrongly received or completely missing.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Bayesian Modeling and Causal Inference · Domain Adaptation and Few-Shot Learning · Anomaly Detection Techniques and Applications
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Energy Technologies and Renewable Energy Sources, Research Centre of Portici, ENEA, 80055 Portici, Italy
Dipartimento di Ingegneria, Università degli Studi della Campania “Luigi Vanvitelli”, Via Roma, 81031 Aversa, Italy
CNIT-Consorzio Nazionale Interuniversitario per le Telecomunicazioni, Complesso Universitario di Monte S. Angelo, Edificio Centri Comuni, Via Cintia, 80126 Naples, Italy
Department of Energy Technologies and Renewable Energy Sources, Research Centre of Casaccia, ENEA, Via Anguillarese, S. Maria di Galeria, 00123 Rome, Italy
Funding
POR Regione Campania
References (34)
Show source-provided references
1. Raol, J.R. (2017). Data Fusion Mathematics: Theory and Practice, CRC Press.
2. Sensor fusion potential exploitation-innovative architectures and illustrative applications10.1109/5.554206 ↗
3. Frontex (2012). Best Practice Operational Guidelines for Automated Border Control (ABC) Systems, European Agency. Available online: https://www.scribd.com/document/169819829/Best-Practice-Operational-Guidelines-for-Automated-Border-Control.
4. Xu, Y., Xu, Y., Lv, T., Cui, L., Wei, F., Wang, G., Lu, Y., Florencio, D., Zhang, C., and Che, W. (2020). LayoutLMv2: Multi-Modal Pre-Training for Visually-Rich Document Understanding. arXiv.10.18653/v1/2021.acl-long.201 ↗
5. Bakkali, S., Ming, Z., Coustaty, M., and Rusiñol, M. (2020, January 25–28). Cross-Modal Deep Networks For Document Image Classification. Proceedings of the 2020 IEEE International Conference on Image Processing (ICIP), Abu Dhabi, United Arab Emirates.10.1109/ICIP40778.2020.9191268 ↗
6. Wang, Z., Wang, C., Zhang, H., Duan, Z., Zhou, M., and Chen, B. (2020, January 26–28). Learning Dynamic Hierarchical Topic Graph with Graph Convolutional Network for Document Classification. Proceedings of the Twenty Third International Conference on Artificial Intelligence and Statistics (AISTAT) 2020, Online.
7. Ullah, I., Manzo, M., Shah, M., and Madden, M. (2019). Graph Convolutional Networks: analysis, improvements and results. arXiv.
8. Simon, M., Rodner, E., and Denzler, J. (2015, January 18–22). Fine-grained classification of identity document types with only one example. Proceedings of the 2015 14th IAPR International Conference on Machine Vision Applications (MVA), Tokyo, Japan.10.1109/MVA.2015.7153149 ↗
9. Sicre, R., Montaser Awal, A., and Furon, T. (2017, January 11–15). Identity documents classification as an image classification problem. Proceedings of the ICIAP 2017—19th International Conference on Image Analysis and Processing, Catania, Italy.10.1007/978-3-319-68548-9_55 ↗
10. Awal, A.M., Ghanmi, N., Sicre, R., and Furon, T. (2017, January 9–15). Complex Document Classification and Localization Application on Identity Document Images. Proceedings of the 2017 14th IAPR International Conference on Document Analysis and Recognition (ICDAR), Kyoto, Japan.10.1109/ICDAR.2017.77 ↗
11. Vilàs Mari, P. (2018). Classification of Identity Documents Using a Deep Convolutional Neural Network. [Master’s Thesis, Universitat Oberta de Catalunya]. Available online: http://hdl.handle.net/10609/73186.
12. Ellena, F. (2018). Deep Convolutional Neural Networks for Document Classification. [Master’s Thesis, Politecnico di Torino]. Available online: http://webthesis.biblio.polito.it/id/eprint/7603.
13. A Comparison of Algorithms for Learning Hidden Variables in Bayesian Factor Graphs in Reduced Normal Form10.1109/TNNLS.2015.2477379 ↗
14. Shi, X., and Manduchi, R. (2003, January 16–22). A Study on Bayes Feature Fusion for Image Classification. Proceedings of the 2003 Conference on Computer Vision and Pattern Recognition Workshop, Madison, WI, USA.10.1109/CVPRW.2003.10090 ↗
15. Olascoaga, L.I.G., Meert, W., Bruyninckx, H., and Verhelst, M. (2016, January 30). Extending Naive Bayes with Precision-Tunable Feature Variables for Resource-Efficient Sensor Fusion. Proceedings of the 2nd Workshop on Artificial Intelligence and Internet of Things, Hague, The Netherlands. AI-IoT@ECAI: 2016.
16. Jin, Y., and Geman, S. (2006, January 17–22). Context and Hierarchy in a Probabilistic Image Model. Proceedings of the 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR’06), New York, NY, USA.
17. A Tree-Based Context Model for Object Recognition10.1109/TPAMI.2011.119 ↗
18. Esposito, A., Faundez-Zanuy, M., Morabito, F.C., and Pasero, E. (2019). Context Analysis Using a Bayesian Normal Graph. Neural Advances in Processing Nonlinear Dynamic Signals, Springer International Publishing.10.1007/978-3-319-95098-3 ↗
19. Redmon, J., and Farhadi, A. (2018). YOLOv3: An Incremental Improvement. arXiv.
20. Redmon, J., and Farhadi, A. (2017, January 21–26). YOLO9000: Better, Faster, Stronger. Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA.10.1109/CVPR.2017.690 ↗
21. Lin, T., Dollár, P., Girshick, R., He, K., Hariharan, B., and Belongie, S. (2017, January 21–26). Feature Pyramid Networks for Object Detection. Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA.10.1109/CVPR.2017.106 ↗
22. Yang, S., Luo, P., Loy, C.C., and Tang, X. (2016, January 27–30). WIDER FACE: A Face Detection Benchmark. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA.10.1109/CVPR.2016.596 ↗
23. Nguyen, T. (2020, December 04). Yoloface. Available online: https://github.com/sthanhng/yoloface.
24. Zhou, X., Yao, C., Wen, H., Wang, Y., Zhou, S., He, W., and Liang, J. (2017, January 21–26). EAST: An Efficient and Accurate Scene Text Detector. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA.10.1109/CVPR.2017.283 ↗
25. Argman (2020, December 04). East. Available online: https://github.com/argman/EAST.
26. Rosebrock, A. (2020, December 04). Detecting Barcodes in Images with Python and OpenCV. Available online: https://www.pyimagesearch.com/2014/11/24/detecting-barcodes-images-python-opencv/.
27. Bassis, S., Esposito, A., and Morabito, F.C. (2015). Simulink Implementation of Belief Propagation in Normal Factor Graphs. Advances in Neural Networks: Computational and Theoretical Issues, Springer International Publishing.
28. Buonanno, A., and Palmieri, F.A.N. (2015, January 12–17). Two-Dimensional Multi-layer Factor Graphs in Reduced Normal Form. Proceedings of the International Joint Conference on Neural Networks, IJCNN2015, Killarney, Ireland.10.1109/IJCNN.2015.7280415 ↗
29. Koller, D., and Friedman, N. (2009). Probabilistic Graphical Models: Principles and Techniques—Adaptive Computation and Machine Learning, The MIT Press.
30. Di Gennaro, G., Buonanno, A., and Palmieri, F.A.N. (2018, January 17–20). Computational Optimization for Normal Form Realization of Bayesian Model Graphs. Proceedings of the 2018 IEEE 28th International Workshop on Machine Learning for Signal Processing (MLSP), Aalborg, Denmark.10.1109/MLSP.2018.8517090 ↗
31. Cover, T.M., and Thomas, J.A. (2006). Elements of Information Theory, Wiley.10.1002/047174882X ↗
32. MIDV-500: a dataset for identity document analysis and recognition on mobile devices in video stream10.18287/2412-6179-2019-43-5-818-824 ↗
33. Harley, A.W., Ufkes, A., and Derpanis, K.G. (2015, January 23–26). Evaluation of deep convolutional nets for document image classification and retrieval. Proceedings of the 2015 13th International Conference on Document Analysis and Recognition (ICDAR), Nancy, France.10.1109/ICDAR.2015.7333910 ↗
34. A systematic analysis of performance measures for classification tasks10.1016/j.ipm.2009.03.002 ↗