cleaver
发表于 2025-3-25 04:44:56
http://reply.papertrans.cn/29/2824/282308/282308_21.png
GUISE
发表于 2025-3-25 10:00:32
http://reply.papertrans.cn/29/2824/282308/282308_22.png
高度表
发表于 2025-3-25 14:40:21
Accelerating Transformer-Based Scene Text Detection and Recognition via Token Pruning image features, which significantly reduces the number of tokens at the input of the transformer. Experimental results show how our network is able to obtain competitive results on multiple public benchmarks while running at significantly higher speeds.
聚集
发表于 2025-3-25 18:32:02
Text Enhancement: Scene Text Recognition in Hazy Weatherent module and a recognition module. Extensive experiments on synthetic and real-world benchmarks demonstrate that our approach significantly improves the performance of popular text recognizers in adverse weather conditions.
ONYM
发表于 2025-3-25 23:24:36
https://doi.org/10.1007/978-3-031-41731-3Document Analysis Systems; artificial intelligence; Document Layout and Parsing; Document Information E
条约
发表于 2025-3-26 03:13:45
http://reply.papertrans.cn/29/2824/282308/282308_26.png
逢迎春日
发表于 2025-3-26 07:06:23
Lecture Notes in Computer Sciencehttp://image.papertrans.cn/e/image/282308.jpg
Microaneurysm
发表于 2025-3-26 09:18:08
https://doi.org/10.1007/978-3-030-30211-5tion of domain specific layout structures, and is further exacerbated by real-world image degradations such as perspective distortions. We propose a lightweight, scalable and generalizable approach to identify text reading order with a multi-modal, multi-task graph convolutional network (GCN) runnin
Perennial长期的
发表于 2025-3-26 14:37:13
http://reply.papertrans.cn/29/2824/282308/282308_29.png
tackle
发表于 2025-3-26 17:05:24
Advances in Global Change Researchattention due to its essential role in various applications like understanding receipts, goods, and traffic signs. However, as existing benchmark datasets for VIE mainly consist of document images without the adequate diversity of layout structures, background disturbs, and entity categories, they c