Scholarlink

The field of image-to-image translation, particularly in the context of denoising historical text documents, has witnessed significant advancements driven by the application of Generative Adversarial Networks (GANs). GANs have shown promise in various image restoration tasks, including unpaired image-to-image translation and denoising. This paper explores and contributes to the existing body of knowledge on denoising historical text documents using GANs, with a focus on addressing specific challenges such as stains, faded ink, background noise, and various types of artificial noise. The field of image-to-image translation, particularly in the context of denoising historical text documents, has witnessed significant advancements driven by the application of Generative Adversarial Networks (GANs). GANs have shown promise in various image restoration tasks, including unpaired image-to-image translation and denoising. This paper explores and contributes to the existing body of knowledge on denoising historical text documents using GANs, with a focus on addressing specific challenges such as stains, faded ink, background noise, and various types of artificial noise.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Denoising Historical Text Documents Using Generative Adversarial Networks

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • P. Preethi,
  • Pradhyumna Upadhya,
  • M. C. Likith,
  • N. Meghana,
  • Shruti Karande,
  • Shreya Gunnan Ramkumar

摘要

The field of image-to-image translation, particularly in the context of denoising historical text documents, has witnessed significant advancements driven by the application of Generative Adversarial Networks (GANs). GANs have shown promise in various image restoration tasks, including unpaired image-to-image translation and denoising. This paper explores and contributes to the existing body of knowledge on denoising historical text documents using GANs, with a focus on addressing specific challenges such as stains, faded ink, background noise, and various types of artificial noise. The field of image-to-image translation, particularly in the context of denoising historical text documents, has witnessed significant advancements driven by the application of Generative Adversarial Networks (GANs). GANs have shown promise in various image restoration tasks, including unpaired image-to-image translation and denoising. This paper explores and contributes to the existing body of knowledge on denoising historical text documents using GANs, with a focus on addressing specific challenges such as stains, faded ink, background noise, and various types of artificial noise.