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 1909.07549cs.CV
收藏

STELA:一个具有学习锚点的实时的场景文本检测器

STELA: A Real-Time Scene Text Detector with Learned Anchor

UESTC(电子科技大学);Sichuan University(四川大学)

Linjie Deng, Yanxiang Gong, Xinchen Lu, Yi Lin, Zheng Ma, Mei Xie

摘要:

为了实现目标框的高覆盖率,传统的基于锚点的检测器的常规策略是在每个空间位置利用多个先验,特别是在场景文本检测任务中。在这项工作中,我们提出了一个简单而直观的方法,面向多个文本检测的特征图的每个位置只与一个参考框。这个想法的灵感来自两阶段R-CNN框架,该框架可以通过使用学习的建议来估计任何形状的对象的位置。我们的方法的目的是将这种机制集成到一个onestage检测器中,并使用通过回归操作获得的学习锚来取代原始锚到最终预测中。基于RetinaNet,该方法在多个公共基准测试中取得了有竞争力的性能,具有完全实时的【形】效率(800 p时为26:5 fps),超过了所有基于锚点的场景文本检测器。此外,由于对锚点设计的关注较少,我们相信我们的方法很容易应用于其他类似的检测任务。该代码将在https://github.com/xhzdeng/stela上公开发布。

英文摘要:

To achieve high coverage of target boxes, a normal strategy of conventional one-stage anchor-based detectors is to utilize multiple priors at each spatial position, especially in scene text detection tasks. In this work, we present a simple and intuitive method for multi-oriented text detection where each location of feature maps only associates with one reference box. The idea is inspired from the twostage R-CNN framework that can estimate the location of objects with any shape by using learned proposals. The aim of our method is to integrate this mechanism into a onestage detector and employ the learned anchor which is obtained through a regression operation to replace the original one into the final predictions. Based on RetinaNet, our method achieves competitive performances on several public benchmarks with a totally real-time efficiency (26:5fps at 800p), which surpasses all of anchor-based scene text detectors. In addition, with less attention on anchor design, we believe our method is easy to be applied on other analogous detection tasks. The code will publicly available at https://github.com/xhzdeng/stela.

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。

↑