TU Dortmund Department of Computer Science LS XII Pattern Recognition GroupPublications → Publication Details

Annotation-free Learning of Deep Representations for Word Spotting using Synthetic Data and Self Labeling


Fabian Wolf and Gernot A. Fink
Proc. Int. Workshop on Document Analysis Systems, pages 293-308, Wuhan, China (virtual), 2020, Winner of the Nakano Best Paper Award.

Word spotting is a popular tool for supporting the first exploration of historic, handwritten document collections. Today, the best performing methods rely on machine learning techniques, which require a high amount of annotated training material. As training data is usually not available in the application scenario, annotation-free methods aim at solving the retrieval task without representative training samples. In this work, we present an annotation-free method that still employs machine learning techniques and therefore outperforms other learning-free approaches. The weakly supervised training scheme relies on a lexicon, that does not need to precisely fit the dataset. In combination with a confidence based selection of pseudo-labeled training samples, we achieve state-of-the-art query-by-example performances. Furthermore, our method allows to perform query-by-string, which is usually not the case for other annotation-free methods.

 [bib] [pdf] [http]