PaperWeekly 03月13日
ICDAR 2025赛事:MapText-25历史地图文字识别竞赛开启报名!
index_new5.html
../../../zaker_core/zaker_tpl_static/wap/tpl_guoji1.html

 

ICDAR 2025历史地图文本检测、识别与链接竞赛测试集现已发布!本次竞赛旨在解决数字化历史地图中,文本信息难以搜索的问题。相比于之前的竞赛,本次竞赛增加了法语土地登记数据集的扩展数据,新增了中国台湾地图数据集,发布了每个数据集的合成训练数据,并改进了评估指标。参赛者将有机会在ICDAR 2025大会上展示成果,并与领域内的研究者交流。

🗺️ICDAR 2025历史地图文本检测、识别与链接竞赛旨在解决数字化历史地图中,文本信息难以搜索的问题,因为这些地图通常以栅格格式存储。

🖋️本次竞赛专注于检测和识别文本信息,例如地名,并将单词链接成地点短语。历史地图文本提取面临着密集文本区域、旋转和弯曲文本以及字符间距过大等挑战。

🇨🇳竞赛引入了多个新特性,包括法语土地登记数据集的扩展数据,新增的来自中央研究院GIS中心的中国台湾地图数据集,其中包含中文字符,以及为每个数据集发布的合成训练数据。

🏆顶级团队将被邀请合作撰写官方竞赛报告,并在ICDAR 2025大会上获得表彰和奖励。这是一个与其他领先研究人员交流和展示成果的机会。

让你更懂AI的 2025-03-13 12:32 北京



We’re excited to announce that the test set for the ICDAR 2025 Competition on Historical Map Text Detection, Recognition, and Linking is now available! 


With just a month to go before the April 1, 2025 deadline, now is the time to put your methods to the test and showcase your innovations in historical map text recognition.



Description


Text on digitized historical maps contains valuable information providing georeferenced political and cultural context, yet the wealth of information in digitized historical maps remains largely inaccessible due to their unsearchable raster format. This competition aims to address the unique challenges of detecting and recognizing textual information (e.g., place names) and linking words to form location phrases. 


While the detection and recognition tasks share similarities with the long line of prior robust reading competitions [1,2], historical map text extraction presents challenges such as dense text regions, rotated and curved text and widely spaced characters which are not very common in scene text extraction problems. The word linking task, in particular, is quite challenging as words can be highly spaced with complicated text-like distractors, even other words appearing between the characters. Furthermore, words within a single location phrase may be divided across multiple lines to optimize label placement. The figure below illustrates primary challenges. 


This 2025 edition is the continuation of the successful 2024 edition. This edition introduces several new features: 

1. expanded data for the French Land Registers dataset 

2. addition of a new Taiwanese maps dataset from the GIS Center at Academia Sinica, featuring Chinese characters 

3. release of synthetic training data for each dataset 

4. improved evaluation metrics and indicators for better feedback



Why Participate?



Top teams will be invited to co-author the official competition report published at ICDAR 2025.

Winners will be recognized on stage at the main ICDAR conference and receive an award.

Benchmark your approach against other leading researchers in the field.



Key Updates


Test set available! Download all three subsets from the competition platform. 


Submissions open! Submit your results and method descriptions via the platform. 


Final deadline: April 1, 2025. 


This is your chance to contribute to cutting-edge research in historical map text detection and recognition! Don't miss out—download the test set and get started today! 


Competition page & details: 

https://rrc.cvc.uab.es/?ch=32


Stay updated: 

https://groups.google.com/g/icdar25-maptext-news 


Looking forward to seeing your groundbreaking contributions to this exciting challenge!

起探讨前沿技术,分享你的研究吧!



?


现在,在「知乎」也能找到我们了

进入知乎首页搜索「PaperWeekly」

点击「关注」订阅我们的专栏吧


·

·

·


阅读原文

跳转微信打开

Fish AI Reader

Fish AI Reader

AI辅助创作,多种专业模板,深度分析,高质量内容生成。从观点提取到深度思考,FishAI为您提供全方位的创作支持。新版本引入自定义参数,让您的创作更加个性化和精准。

FishAI

FishAI

鱼阅,AI 时代的下一个智能信息助手,助你摆脱信息焦虑

联系邮箱 441953276@qq.com

相关标签

ICDAR 2025 历史地图 文本检测 文本识别
相关文章