Koch T, Liebezeit F, Rieß C, Christlein V, Köhler T (2022)
Publication Language: English
Publication Type: Conference contribution, Conference Contribution
Publication year: 2022
Publisher: IEEE
Series: Pattern Recognition
Pages Range: 2792-2799
Conference Proceedings Title: 2022 26th International Conference on Pattern Recognition (ICPR)
ISBN: 978-1-6654-9062-7
URI: https://faui1-files.cs.fau.de/public/publications/mmsec/2022-Koch-ICPR.pdf
DOI: 10.1109/ICPR56361.2022.9956423
Dynamic environments require adaptive applications. One particular machine learning problem in dynamic environments is open world recognition. It characterizes a continuously changing domain where only some classes are seen in one batch of the training data and such batches can only be learned incrementally. Open world recognition is a demanding task that is, to the best of our knowledge, addressed by only a few methods. This work introduces a modification of the widely known Extreme Value Machine (EVM) to enable open world recognition. Our proposed method extends the EVM with a partial model fitting function by neglecting unaffected space during an update. This reduces the training time by a factor of 28. In addition, we provide a modified model reduction using weighted maximum K-set cover to strictly bound the model complexity and reduce the computational effort by a factor of 3.5 from 2.1 s to 0.6 s. In our experiments, we rigorously evaluate openness with two novel evaluation protocols. The proposed method achieves superior accuracy of about 12 % and computational efficiency in the tasks of image classification and face recognition.
APA:
Koch, T., Liebezeit, F., Rieß, C., Christlein, V., & Köhler, T. (2022). Exploring the Open World Using Incremental Extreme Value Machines. In 2022 26th International Conference on Pattern Recognition (ICPR) (pp. 2792-2799). Montréal, CA: IEEE.
MLA:
Koch, Tobias, et al. "Exploring the Open World Using Incremental Extreme Value Machines." Proceedings of the International Conference on Pattern Recognition (ICPR), Montréal IEEE, 2022. 2792-2799.
BibTeX: Download