存心 发表于 2025-3-28 16:52:36

http://reply.papertrans.cn/29/2824/282318/282318_41.png

使乳化 发表于 2025-3-28 21:13:59

Paula Kasares,Ane Ortega,Estibaliz Amorrortuition at ICPR 2022. The data set includes 15 different chart categories, including 22,923 training images and 13,260 test images. We have implemented a vision-based transformer model that produces state-of-the-art results in chart classification.

collagen 发表于 2025-3-28 23:57:09

http://reply.papertrans.cn/29/2824/282318/282318_43.png

contradict 发表于 2025-3-29 04:57:04

Karel van Hulle,Leo van der Tasents..The evaluations were carried out on the KABOOM-ONOMATOPOEIA dataset and show the relevance of our method in comparison with methods of the literature, which makes it a promising tool in the field of scene text detection.

出血 发表于 2025-3-29 09:38:02

http://reply.papertrans.cn/29/2824/282318/282318_45.png

Trochlea 发表于 2025-3-29 15:11:03

Document Analysis and Recognition – ICDAR 2023 WorkshopsSan José, CA, USA, A

猛烈责骂 发表于 2025-3-29 15:52:02

Beyond Human Forgeries: An Investigation into Detecting Diffusion-Generated Handwritingn methods. Our experiments indicate that the strongest discriminative features do not come from generation artifacts, letter shapes, or the generative model’s architecture, but instead originate from real-world artifacts in genuine handwriting that are not reproduced by generative methods.

osteoclasts 发表于 2025-3-29 20:27:46

Leveraging Large Language Models for Topic Classification in the Domain of Public Affairstokens. Our experiments assess the performance of 4 different Spanish LLMs to classify up to 30 different topics in the data in different configurations. The results shows that LLMs can be of great use to process domain-specific documents, such as those in the domain of public affairs.

jarring 发表于 2025-3-30 00:22:57

Using GANs for Domain Adaptive High Resolution Synthetic Document Generationbased) represented a preliminary step towards the generation of realistic document images, it was impeded by its incapacity to produce high-resolution outputs. Our research aims to overcome this restriction, enhancing the DocSynth model’s capacity to generate high-resolution document images. Additio

Mercantile 发表于 2025-3-30 06:39:17

A Survey and Approach to Chart Classificationition at ICPR 2022. The data set includes 15 different chart categories, including 22,923 training images and 13,260 test images. We have implemented a vision-based transformer model that produces state-of-the-art results in chart classification.
页: 1 2 3 4 [5] 6
查看完整版本: Titlebook: Document Analysis and Recognition – ICDAR 2023 Workshops; San José, CA, USA, A Mickael Coustaty,Alicia Fornés Conference proceedings 2023 T