No abstract

Understanding document images (e.g., invoices) has been an important research topic and has many applications in document processing automation. Through the latest advances in deep learning-based Optical Character Recognition (OCR), current Visual Document Understanding (VDU) systems have come to be designed based on OCR. Although such OCR-based approach promise reasonable performance, they suffer from critical problems induced by the OCR, e.g., (1) expensive computational costs and (2) performance degradation due to the OCR error propagation. In this paper, we propose a novel VDU model that is end-to-end trainable without underpinning OCR framework. To this end, we propose a new task and a synthetic document image generator to pre-train the model to mitigate the dependencies on largescale real document images. Our approach achieves state-of-the-art performance on various document understanding tasks in public benchmark datasets and private industrial service datasets. Through extensive experiments and analysis, we demonstrate the effectiveness of the proposed model especially with consideration for a real-world application.

Biomolecular condensates are subcellular organizations where functionally related proteins and nucleic acids are assembled through liquid–liquid phase separation, allowing them to develop on a larger scale without a membrane. However, biomolecular condensates are highly vulnerable to disruptions from genetic risks and various factors inside and outside the cell and are strongly implicated in the pathogenesis of many neurodegenerative diseases. In addition to the classical view of the nucleation-polymerization process that triggers the protein aggregation from the misfolded seed, the pathologic transition of biomolecular condensates can also promote the aggregation of proteins found in the deposits of neurodegenerative diseases. Furthermore, it has been suggested that several protein or protein-RNA complexes located in the synapse and along the neuronal process are neuron-specific condensates displaying liquid-like properties. As their compositional and functional modifications play a crucial role in the context of neurodegeneration, further research is needed to fully understand the role of neuronal biomolecular condensates. In this article, we will discuss recent findings that explore the pivotal role of biomolecular condensates in the development of neuronal defects and neurodegeneration.

Text localization from the digital image is the first step for the optical character recognition task. Conventional image processing based text localization performs adequately for specific examples. Yet, a general text localization are only archived by recent deep-learning based modalities. Here we present document Text Localization Generative Adversarial Nets (TLGAN) which are deep neural networks to perform the text localization from digital image. TLGAN is an versatile and easy-train text localization model requiring a small amount of data. Training only ten labeled receipt images from Robust Reading Challenge on Scanned Receipts OCR and Information Extraction (SROIE), TLGAN achieved 99.83% precision and 99.64% recall for SROIE test data. Our TLGAN is a practical text localization solution requiring minimal effort for data labeling and model training and producing a state-of-art performance.