改进 LLM 的文件解析
layout-parsing 开源项目推荐(6 个)
layout-parsing 话题在 GitHub 上没有填写简介,这里按 Star 数汇总最相关的热门仓库。
热门项目 全部 6
GitHub 共 6 个仓库,已全部展示Visual Scene Graph for PowerPoint. Extracts structured layout data from PPTX files, enabling AI agents to parse element positions, understand spatial relationships, and detect layout patterns. Exposes bounding boxes, z-order, and spatial relations that humans see but machines couldn’t access. Built on python-pptx. MIT licensed.
A lightweight Python library for metadata-rich document chunking in Retrieval-Augmented Generation (RAG) workflows. It leverages Azure AI Document Intelligence to enhance chunking by retaining hierarchical structure, page numbers, and bounding boxes for seamless integration with PDF viewers.
A Flask microservice for extracting structured data from PDFs and images using PaddleOCR layout parsing API.
--UNDER CONSTRUCTION-- (Undergrad Research) Exploring layout parsing capabilities in Python
Scale-to-zero dots.mocr OCR and layout parsing API on Modal (vLLM + FastAPI), with an editable Word exporter and a browser playground
关于这个话题
- 清单按 Star 数挑选该话题下最热门的 6 个项目,对应的话题主页见 https://github.com/topics/layout-parsing。
- 该话题在 GitHub 上共有 6 个仓库, 这里呈现的是其中最具代表性的 6 个,不是全量列表。
- 星标与复刻数以 GitHub 上的实时数字为准。
- 当前 6 个项目累计 3159 Star,覆盖 2 种主要语言。
- 中文简介由本站编译整理,可点「查看原文」对照项目方提供的原始说明。