Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wang, Junxiao, Zhang, Ting, Yu, Heng, Wang, Jingdong, Huang, Hua
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2502.13855
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915161257803776
author Wang, Junxiao
Zhang, Ting
Yu, Heng
Wang, Jingdong
Huang, Hua
author_facet Wang, Junxiao
Zhang, Ting
Yu, Heng
Wang, Jingdong
Huang, Hua
contents Geometric diagrams are critical in conveying mathematical and scientific concepts, yet traditional diagram generation methods are often manual and resource-intensive. While text-to-image generation has made strides in photorealistic imagery, creating accurate geometric diagrams remains a challenge due to the need for precise spatial relationships and the scarcity of geometry-specific datasets. This paper presents MagicGeo, a training-free framework for generating geometric diagrams from textual descriptions. MagicGeo formulates the diagram generation process as a coordinate optimization problem, ensuring geometric correctness through a formal language solver, and then employs coordinate-aware generation. The framework leverages the strong language translation capability of large language models, while formal mathematical solving ensures geometric correctness. We further introduce MagicGeoBench, a benchmark dataset of 220 geometric diagram descriptions, and demonstrate that MagicGeo outperforms current methods in both qualitative and quantitative evaluations. This work provides a scalable, accurate solution for automated diagram generation, with significant implications for educational and academic applications.
format Preprint
id arxiv_https___arxiv_org_abs_2502_13855
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MagicGeo: Training-Free Text-Guided Geometric Diagram Generation
Wang, Junxiao
Zhang, Ting
Yu, Heng
Wang, Jingdong
Huang, Hua
Computer Vision and Pattern Recognition
Geometric diagrams are critical in conveying mathematical and scientific concepts, yet traditional diagram generation methods are often manual and resource-intensive. While text-to-image generation has made strides in photorealistic imagery, creating accurate geometric diagrams remains a challenge due to the need for precise spatial relationships and the scarcity of geometry-specific datasets. This paper presents MagicGeo, a training-free framework for generating geometric diagrams from textual descriptions. MagicGeo formulates the diagram generation process as a coordinate optimization problem, ensuring geometric correctness through a formal language solver, and then employs coordinate-aware generation. The framework leverages the strong language translation capability of large language models, while formal mathematical solving ensures geometric correctness. We further introduce MagicGeoBench, a benchmark dataset of 220 geometric diagram descriptions, and demonstrate that MagicGeo outperforms current methods in both qualitative and quantitative evaluations. This work provides a scalable, accurate solution for automated diagram generation, with significant implications for educational and academic applications.
title MagicGeo: Training-Free Text-Guided Geometric Diagram Generation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2502.13855