RT-DocLayout achieves real-time document analysis with unified architecture

Researchers have developed RT-DocLayout, an efficient end-to-end framework for document layout analysis and reading order prediction. This single model, built on RT-DETR, unifies classification, detection, segmentation, and reading order prediction within a 33M-parameter architecture. Experiments show RT-DocLayout achieves state-of-the-art performance with real-time inference speeds and significantly improves downstream OCR engine reconstruction quality. AI

IMPACT This model could significantly improve the efficiency and accuracy of document parsing and information extraction systems.

RANK_REASON The item is a research paper detailing a new model and its performance on benchmarks. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• alphaXiv
• arXiv
• CatalyzeX Code Finder for Papers
• Connected Papers
• CORE Recommender
• DagsHub
• Gotit.pub
• Hugging Face
• Influence Flower
• Litmaps
• RT-DETR
• RT-DocLayout
• ScienceCast
• scite Smart Citations

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →