用Transformer生成更连贯的手写笔画，提升识别准确率。
TrInk: Ink Generation with Transformer Network
- 基于Transformer架构，引入缩放位置编码与高斯记忆掩码增强文本-笔画对齐
- 在IAM-OnDB数据集上，字符错误率降低35.56%，词错误率降低29.66%
- 适合关注手写生成质量与可读性的研究者与应用开发者
本文提出TrInk，一种基于Transformer的墨水生成模型，能有效捕捉全局依赖关系。为更好实现输入文本与生成笔画点之间的对齐，我们在交叉注意力模块中引入缩放位置编码和高斯记忆掩码。此外，设计了主观与客观评估流程，全面衡量生成手写的可读性与风格一致性。实验表明，该模型在IAM-OnDB数据集上相较先前方法，字符错误率（CER）降低35.56%，词错误率（WER）降低29.66%。我们提供了包含TrInk及基线模型手写样本的演示页面：https://akahello-a11y.github.io/trink-demo/
原文摘要 · Abstract (English)
In this paper, we propose TrInk, a Transformer-based model for ink generation, which effectively captures global dependencies. To better facilitate the alignment between the input text and generated stroke points, we introduce scaled positional embeddings and a Gaussian memory mask in the cross-attention module. Additionally, we design both subjective and objective evaluation pipelines to comprehensively assess the legibility and style consistency of the generated handwriting. Experiments demonstrate that our Transformer-based model achieves a 35.56\% reduction in character error rate (CER) and an 29.66% reduction in word error rate (WER) on the IAM-OnDB dataset compared to previous methods. We provide an demo page with handwriting samples from TrInk and baseline models at: https://akahello-a11y.github.io/trink-demo/
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。