Christlein V, Maier A (2018)
Publication Type: Conference contribution
Publication year: 2018
Publisher: Institute of Electrical and Electronics Engineers Inc.
Edited Volumes: Proceedings - 13th IAPR International Workshop on Document Analysis Systems, DAS 2018
Pages Range: 169-174
Conference Proceedings Title: 2018 13th IAPR International Workshop on Document Analysis Systems (DAS)
Event location: Vienna, Austria
ISBN: 9781538633465
URI: https://www5.informatik.uni-erlangen.de/Forschung/Publikationen/2018/Christlein18-ECA.pdf
DOI: 10.1109/DAS.2018.9
The encoding of local features is an essential part for writer identification and writer retrieval. While CNN activations have already been used as local features in related works, the encoding of these features has attracted little attention so far. In this work, we compare the established VLAD encoding with triangulation embedding. We further investigate generalized max pooling as an alternative to sum pooling and the impact of decorrelation and Exemplar SVMs. With these techniques, we set new standards on two publicly available datasets (ICDAR13, KHATT).
APA:
Christlein, V., & Maier, A. (2018). Encoding CNN Activations for Writer Recognition. In IEEE (Eds.), 2018 13th IAPR International Workshop on Document Analysis Systems (DAS) (pp. 169-174). Vienna, Austria: Institute of Electrical and Electronics Engineers Inc..
MLA:
Christlein, Vincent, and Andreas Maier. "Encoding CNN Activations for Writer Recognition." Proceedings of the 13th IAPR International Workshop on Document Analysis Systems (DAS), Vienna, Austria Ed. IEEE, Institute of Electrical and Electronics Engineers Inc., 2018. 169-174.
BibTeX: Download