arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2407.05562cs.CV
收藏

聚焦完整字符:用于场景文本识别的判别性字符建模

Focus on the Whole Character: Discriminative Character Modeling for Scene Text Recognition

University of Science and Technology of China(中国科学技术大学)

Bangbang Zhou, Yadong Qu, Zixiao Wang, Zicheng Li, Boqiang Zhang, Hongtao Xie

AI总结:

针对场景文本中严重变形字符导致的类内方差大、类间方差小问题,提出CACE编码器与I^2CL损失以增强字符特征判别性,并在多个基准上取得SOTA结果。

摘要:

近年来,场景文本识别(STR)模型在性能上取得了显著提升。然而,现有模型在识别涉及严重变形和透视字符等因素的挑战性文本时仍面临困难。这些挑战性文本主要导致两个问题:(1)较大的类内方差;(2)较小的类间方差。一个极度变形的字符在视觉上可能与同一类别中的其他字符存在明显差异,而不同类别字符之间的方差却相对较小。为解决上述问题,我们提出了一种新颖方法,通过丰富字符特征来增强字符的可判别性。首先,我们提出了由多个模块堆叠而成的字符感知约束编码器(Character-Aware Constraint Encoder,CACE)。CACE在每个模块中引入衰减矩阵,以显式引导每个 token 的注意区域。通过持续使用衰减矩阵,CACE使 token 能够感知字符级别的形态信息。其次,我们引入了类内-类间一致性损失(Intra-Inter Consistency Loss,I^2CL),以在特征空间中同时考虑类内紧凑性和类间可分性。I^2CL通过为每个字符类别学习一个长期记忆单元来提升特征的判别能力。在使用合成数据进行训练后,我们的模型在常用基准测试上达到了最先进性能(准确率94.1%),并在 Union14M-Benchmark 上达到61.6%的准确率。代码可在 https://github.com/bang123-box/CFE 获取。

英文摘要:

Recently, scene text recognition (STR) models have shown significant performance improvements. However, existing models still encounter difficulties in recognizing challenging texts that involve factors such as severely distorted and perspective characters. These challenging texts mainly cause two problems: (1) Large Intra-Class Variance. (2) Small Inter-Class Variance. An extremely distorted character may prominently differ visually from other characters within the same category, while the variance between characters from different classes is relatively small. To address the above issues, we propose a novel method that enriches the character features to enhance the discriminability of characters. Firstly, we propose the Character-Aware Constraint Encoder (CACE) with multiple blocks stacked. CACE introduces a decay matrix in each block to explicitly guide the attention region for each token. By continuously employing the decay matrix, CACE enables tokens to perceive morphological information at the character level. Secondly, an Intra-Inter Consistency Loss (I^2CL) is introduced to consider intra-class compactness and inter-class separability at feature space. I^2CL improves the discriminative capability of features by learning a long-term memory unit for each character category. Trained with synthetic data, our model achieves state-of-the-art performance on common benchmarks (94.1% accuracy) and Union14M-Benchmark (61.6% accuracy). Code is available at https://github.com/bang123-box/CFE.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑