论文概要
研究领域: CV 作者: Yonatan Haile Medhanie, Yuanhua Ni 发布时间: 2026-04-22 arXiv: 2604.20813
中文摘要
基于Transformer的OCR模型在拉丁文和CJK文字上表现强劲，但其在非洲音节书写系统上的应用仍然有限。我们首次将TrOCR应用于使用吉兹字母的印刷提格里尼亚语。从预训练模型出发，我们将字节级BPE分词器扩展以覆盖230个吉兹字符，并引入词感知损失加权来解决将拉丁中心化BPE约定应用于新文字时出现的系统性词边界失败问题。未修改的模型在吉兹文本上无法产生可用输出。适应后，TrOCR-Printed变体在GLOCR数据集的5000张合成图像测试集上达到0.22%字符错误率和97.20%精确匹配准确率。消融研究确认词感知损失加权是关键组件，仅词汇扩展相比，将CER降低两个数量级。完整流程在单个8GB消费级GPU上训练不到三小时。所有代码、模型权重和评估脚本均已公开发布。
原文摘要
Transformer-based OCR models have shown strong performance on Latin and CJK scripts, but their application to African syllabic writing systems remains limited. We present the first adaptation of TrOCR for printed Tigrinya using the Ge'ez script. Starting from a pre-trained model, we extend the byte-level BPE tokenizer to cover 230 Ge'ez characters and introduce Word-Aware Loss Weighting to resolve systematic word-boundary failures that arise when applying Latin-centric BPE conventions to a new script. The unmodified model produces no usable output on Ge'ez text. After adaptation, the TrOCR-Printed variant achieves 0.22% Character Error Rate and 97.20% exact match accuracy on a held-out test set of 5,000 synthetic images from the GLOCR dataset. An ablation study confirms that Word-Aware Los...
--- *自动采集于 2026-04-24*
#论文 #arXiv #CV #小凯