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 2208.03364cs.CVcs.AI
收藏

GLASS:场景文本识别的全局到局部注意

GLASS: Global to Local Attention for Scene-Text Spotting

Technion(以色列理工学院);AWS AI Labs(亚马逊云科技人工智能实验室)

Roi Ronen, Shahar Tsiper, Oron Anschel, Inbal Lavi, Amir Markovitz, R. Manmatha

摘要:

近年来,文本识别的主要模式是将文本检测和识别的任务结合到一个端到端的框架中。在这种模式下,这两个任务都是通过操作从输入图像中提取的共享全局特征图来完成的。端到端方法面临的主要挑战之一是在识别跨尺度变化(更小或更大的文本)和任意单词旋转角度的文本时性能下降。在这项工作中,我们提出了一种新的全球到本地的注意力机制的文本定位,称为玻璃,融合在一起的全球和本地功能来解决这些挑战。全局特征从共享主干中提取,保留整个图像的上下文信息,而局部特征则在调整大小的高分辨率旋转单词裁剪上单独计算。从当地作物中提取的信息消除了尺度和单词旋转的许多固有困难。我们展示了跨尺度和角度的性能分析,突出了尺度和角度极限的改进。此外,我们引入了一个方向感知的损失项监督检测任务,并显示其在所有角度的检测和识别性能的贡献。最后,我们表明,GLASS是通用的,将其纳入其他领先的文本定位架构,提高其文本定位性能。我们的方法在多个基准测试中取得了最先进的结果,包括新发布的TextOCR。

英文摘要:

In recent years, the dominant paradigm for text spotting is to combine the tasks of text detection and recognition into a single end-to-end framework. Under this paradigm, both tasks are accomplished by operating over a shared global feature map extracted from the input image. Among the main challenges that end-to-end approaches face is the performance degradation when recognizing text across scale variations (smaller or larger text), and arbitrary word rotation angles. In this work, we address these challenges by proposing a novel global-to-local attention mechanism for text spotting, termed GLASS, that fuses together global and local features. The global features are extracted from the shared backbone, preserving contextual information from the entire image, while the local features are computed individually on resized, high-resolution rotated word crops. The information extracted from the local crops alleviates much of the inherent difficulties with scale and word rotation. We show a performance analysis across scales and angles, highlighting improvement over scale and angle extremities. In addition, we introduce an orientation-aware loss term supervising the detection task, and show its contribution to both detection and recognition performance across all angles. Finally, we show that GLASS is general by incorporating it into other leading text spotting architectures, improving their text spotting performance. Our method achieves state-of-the-art results on multiple benchmarks, including the newly released TextOCR.

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。

↑