Annotation-free Learning of Deep Representations for Word Spotting using Synthetic Data and Self Labeling
This provides an annotation-free solution for researchers and archivists exploring historic document collections, though it is incremental as it builds on existing machine learning techniques.
The paper tackles the problem of word spotting in historic handwritten documents without requiring annotated training data by using synthetic data and self-labeling, achieving state-of-the-art query-by-example performance and enabling query-by-string retrieval.
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.