EATEN: Entity-aware Attention for Single Shot Visual Text Extraction
This addresses the need for efficient entity extraction in documents like tickets and passports, but it is incremental as it builds on existing attention-based methods with specific enhancements.
The paper tackles the problem of extracting entities from images in OCR applications by proposing EATEN, an end-to-end trainable system that eliminates post-processing, achieving state-of-the-art performance on a new dataset of 0.6 million images across three real-world scenarios.
Extracting entity from images is a crucial part of many OCR applications, such as entity recognition of cards, invoices, and receipts. Most of the existing works employ classical detection and recognition paradigm. This paper proposes an Entity-aware Attention Text Extraction Network called EATEN, which is an end-to-end trainable system to extract the entities without any post-processing. In the proposed framework, each entity is parsed by its corresponding entity-aware decoder, respectively. Moreover, we innovatively introduce a state transition mechanism which further improves the robustness of entity extraction. In consideration of the absence of public benchmarks, we construct a dataset of almost 0.6 million images in three real-world scenarios (train ticket, passport and business card), which is publicly available at https://github.com/beacandler/EATEN. To the best of our knowledge, EATEN is the first single shot method to extract entities from images. Extensive experiments on these benchmarks demonstrate the state-of-the-art performance of EATEN.