%0 Conference Proceedings
%T Segmentation-Free Alignment of Arbitrary Symbol Transcripts to Images
%A Pau Torras
%A Mohamed Ali Souibgui
%A Sanket Biswas
%A Alicia Fornes
%B Document Analysis and Recognition – ICDAR 2023 Workshops
%D 2023
%V 14193
%F Pau Torras2023
%O DAG
%O exported from refbase (http://refbase.cvc.uab.es/show.php?record=3850), last updated on Mon, 20 Nov 2023 12:00:47 +0100
%X Developing arbitrary symbol recognition systems is a challenging endeavour. Even using content-agnostic architectures such as few-shot models, performance can be substantially improved by providing a number of well-annotated examples into training. In some contexts, transcripts of the symbols are available without any position information associated to them, which enables using line-level recognition architectures. A way of providing this position information to detection-based architectures is finding systems that can align the input symbols with the transcription. In this paper we discuss some symbol alignment techniques that are suitable for low-data scenarios and provide an insight on their perceived strengths and weaknesses. In particular, we study the usage of Connectionist Temporal Classification models, Attention-Based Sequence to Sequence models and we compare them with the results obtained on a few-shot recognition system.
%K Historical Manuscripts
%K Symbol Alignment
%U https://link.springer.com/chapter/10.1007/978-3-031-41498-5_6
%P 83-93