Abstract: Vision Transformer (ViT) has recently emerged as a state-of-the-art approach for text recognition. However, ViT-based text recognition methods remain vulnerable to noise, broken word ...
Abstract: Ancient handwritten Chinese character recognition plays a vital role in advancing the digital preservation of Chinese historical and cultural heritage. In recent years, deep learning-based ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results