Article

Unified Donut Transformer Meets Oriented Object Detection: A Hybrid Framework for Structured Parsing of 2D Engineering Drawings

2025-06-10

Abstract excerpt

The precise and structured extraction of critical technical specifications embedded within complex 2D engineering drawings constitutes a cornerstone for achieving high-fidelity manufacturing outcomes. Conventional manual extraction methodologies are inherently inefficient, susceptible to human error, and ill-suited for handling intricate layouts, while legacy Optical Character Recognition (OCR) systems frequently...

Topics

Open a Topic to create a Post that cites this publication.

Identifiers and source

Literature Corpus work: d7f75b7d-64ff-575e-b642-b2d5fbd10646
DOI: 10.20944/preprints202506.0731.v1

Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 12 closest

Unified Donut Transformer Meets Oriented Object Detection: A Hybrid Framework for Structured Parsing of 2D Engineering DrawingsDOI 10.20944/preprints202506.0731.v1Read

LLM-Integrated Semantic Deep Learning Framework for Automated Floor Plan Analysis, Estimation and Design GuidanceSemantic proximity Read

An Intelligent Deep Learning-Based Monitoring Framework for Additive ManufactureSemantic proximity Read

OCR4all - An Open-Source Tool Providing a (Semi-)Automatic OCR Workflow for Historical PrintingsSemantic proximity Read

LLM-3D Print: Large Language Models To Monitor and Control 3D PrintingSemantic proximity Read

Archaeology drawing generation algorithm based on multi-branch feature cross fusionSemantic proximity Read

Benchmarking and Stress-Testing Off-the-Shelf Vision Embeddings for Surface Re-Identification in Open-Source IntelligenceSemantic proximity Read

From photos to sketches - how humans and deep neural networks process objects across different levels of visual abstractionSemantic proximity Read

Optimal Surface Defect Detector Design Based on Deep Learning for 3D GeometrySemantic proximity Read

A Hybrid Template-Guided Deep Learning Framework for OCR-Oriented Restoration of Degraded Tax DocumentsSemantic proximity Read

StrokePlan-RNN: Predicting Ordered Drawing Procedures from Face Line ArtSemantic proximity Read

Synthetic Meets Authentic: Leveraging Text-to-Image Generated Datasets for Apple Detection in Orchard EnvironmentsSemantic proximity Read

Integrating Coverage-Attention Mechanisms for Improved Chemical Image RecognitionSemantic proximity Read