arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2509.20003cs.CVcs.AIcs.CLcs.LG
收藏

基于主动学习的表格检测

Table Detection with Active Learning

Indian Institute of Technology Jodhpur Rajasthan, India(印度理工学院贾尔普尔拉贾斯坦)

Somraj Gautam, Nachiketa Purohit, Gaurav Harit

AI总结:

本文提出一种结合不确定性与多样性的主动学习方法用于表格检测,在有限标注预算下减少标注成本,并在TableBank数据集上以CascadeTabNet和YOLOv9验证,取得更高mAP。

摘要:

高效的数据标注仍然是机器学习中的一个关键挑战,尤其是在需要大量标注数据的对象检测任务中。主动学习(AL)已成为一种有前景的解决方案,通过选择最具信息量的样本来最小化标注成本。虽然传统的主动学习方法主要依赖于基于不确定性的选择,但最近的进展表明,在对象检测任务中引入基于多样性的策略可以提升采样效率。我们的方法确保选择具有代表性的示例,从而改善模型的泛化能力。我们在两个基准数据集(TableBank-LaTeX、TableBank-Word)上,使用最先进的表格检测架构CascadeTabNet和YOLOv9评估了我们的方法。结果表明,基于主动学习的示例选择显著优于随机采样,在有限预算下减少了标注工作量,同时保持了与全监督模型相当的性能。在相同的标注预算内,我们的方法取得了更高的mAP分数。

英文摘要:

Efficient data annotation remains a critical challenge in machine learning, particularly for object detection tasks requiring extensive labeled data. Active learning (AL) has emerged as a promising solution to minimize annotation costs by selecting the most informative samples. While traditional AL approaches primarily rely on uncertainty-based selection, recent advances suggest that incorporating diversity-based strategies can enhance sampling efficiency in object detection tasks. Our approach ensures the selection of representative examples that improve model generalization. We evaluate our method on two benchmark datasets (TableBank-LaTeX, TableBank-Word) using state-of-the-art table detection architectures, CascadeTabNet and YOLOv9. Our results demonstrate that AL-based example selection significantly outperforms random sampling, reducing annotation effort given a limited budget while maintaining comparable performance to fully supervised models. Our method achieves higher mAP scores within the same annotation budget.

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。

↑