Recognition-free Question Answering on Handwritten Document Collections
This addresses a specific bottleneck in document image analysis for researchers and practitioners dealing with handwritten collections, though it is incremental as it builds on existing recognition-free methods.
The paper tackles the problem of reduced recognition performance in question answering on handwritten documents by proposing a recognition-free approach, which outperforms state-of-the-art models on BenthamQA and HW-SQuAD datasets.
In recent years, considerable progress has been made in the research area of Question Answering (QA) on document images. Current QA approaches from the Document Image Analysis community are mainly focusing on machine-printed documents and perform rather limited on handwriting. This is mainly due to the reduced recognition performance on handwritten documents. To tackle this problem, we propose a recognition-free QA approach, especially designed for handwritten document image collections. We present a robust document retrieval method, as well as two QA models. Our approaches outperform the state-of-the-art recognition-free models on the challenging BenthamQA and HW-SQuAD datasets.