2024年3月4日 · While end-to-end OCR methods offer improved accuracy over layout-based approaches, they often grapple with significant repetition issues, especially with complex layouts in Out-Of-Domain (OOD) this http URL tackle this issue, we propose LOCR, a model that integrates location guiding into the transformer architecture during autoregression.

To tackle this issue, we propose LOCR1, a model that integrates loca- tion guiding into the transformer architecture during autoregression. We train the model on an original large-scale dataset comprising over 53M text-location pairs from 89K academic document pages, including bounding boxes for words, tables and mathematical symbols.

最近为大家介绍了一些自然场景文本检测识别（STR）领域中的检测模型和端到端模型，今天我们要看的是一个识别模型—— GTC：Guided Training of CTC。 这篇文章为2020年2月由南洋理工大学和商汤科技的研究者共同发表…

2025年3月19日 · We train the model on an original large-scale dataset comprising over 53M text-location pairs from 89K academic document pages, including bounding boxes for words, tables and mathematical symbols. LOCR adeptly handles various formatting elements and generates content in Markdown language.

In this paper, we present VisFocus, an OCR-free method designed to better exploit the vision encoder's capacity by coupling it directly with the language prompt. To do so, we replace the down-sampling layers with layers that receive the input prompt and allow highlighting relevant parts of the document, while disregarding others.

2024年6月1日 · 此外，为了促进对文档细粒度理解的研究，作者还打造了一个中英双语的benchmark，已经开源了数据和评测代码，共包含以下9种任务： Page-level OCR Region-level OCR Line-level OCR Color-guided OCR Region-level translation Region-level summary In-document figure caption Multi-page multi-region OCR Cross ...

We introduce LOCR, a location-guided document understanding model, together with an original large-scale dataset and an interactive OCR mode to align with human intention (see Figure 1 for an overview).

本文旨在解决OCR在处理复杂排版的文档时重复率高的问题，提出了一种集成位置引导的transformer架构模型LOCR。 LOCR模型在自回归过程中集成位置引导，能够更好地处理复杂排版的文档，生成Markdown格式的内容，并在多个评价指标上优于现有方法。 本文使用了超过77M个文本-位置对的数据集进行训练，包括单词、表格和数学符号的边界框。 在arXiv数据集上，LOCR将重复率从4.4%降低到0.5%；在OOD量子物理文档和市场营销文档中，LOCR的重 …

We propose a novel dictionary-guided sense text recognition approach that could be used to improve many state-of-the-art models. We also introduce a new benchmark dataset (namely, VinText) for Vietnamese scene text recognition.

与专用OCR模型相比，当前多模态大模型的识字能力相对较弱。 直接使用多模态大模型做视觉信息抽取往往会出现错字。 本项目使用OCR结果来引导多模态大模型输出，以期得到更高的信息抽取准确率。 算法详情见 博客 使用Qwen2VL-2B的例子