OLALA: Object-Level Active Learning Based Layout Annotation
OLALA: Object-Level Active Learning Based Layout Annotation
复制标题
DOI：
--
复制
发表时间：
2020-10
期刊：
影响因子：
--
通讯作者：
Zejiang Shen;Jian Zhao;Melissa Dell;Yaoliang Yu;Weining Li
中科院分区：
文献类型：
--
作者：
Zejiang Shen;Jian Zhao;Melissa Dell;Yaoliang Yu;Weining Li
在布局对象检测问题中，地面真实数据集是通过单独标注对象实例来构造的。然而，对象检测的主动学习通常在图像级别进行，而不是在对象级别进行。由于物体在不同图像中的出现频率不同，图像级别的主动学习可能会受到对常见物体的过度暴露。这降低了人类标记的效率。本文介绍了一个基于对象级主动学习的版面标注框架Olala，它包括一个对象评分方法和一个预测校正算法。目标打分方法综合考虑目标类别和位置，估计目标预测信息量。它只选择图像中最模糊的对象预测区域供注释员标记，优化了注解预算的使用。对于未选择的模型预测，我们提出了一种修正算法，以纠正两类潜在的误差，并从地面事实中进行较少的监督。然后，将人工注释和模型预测对象合并为新的图像注释，用于训练对象检测模型。在模拟标注实验中，我们表明Olala有助于更有效地创建数据集，并且与图像级主动学习基线相比，训练模型的准确率有了显著的提高。代码可在此HTTPS URL中找到
In layout object detection problems, the ground-truth datasets are constructed by annotating object instances individually. Yet active learning for object detection is typically conducted at the image level, not at the object level. Because objects appear with different frequencies across images, image-level active learning may be subject to over-exposure to common objects. This reduces the efficiency of human labeling. This work introduces an Object-Level Active Learning based Layout Annotation framework, OLALA, which includes an object scoring method and a prediction correction algorithm. The object scoring method estimates the object prediction informativeness considering both the object category and the location. It selects only the most ambiguous object prediction regions within an image for annotators to label, optimizing the use of the annotation budget. For the unselected model predictions, we propose a correction algorithm to rectify two types of potential errors with minor supervision from ground-truths. The human annotated and model predicted objects are then merged as new image annotations for training the object detection models. In simulated labeling experiments, we show that OLALA helps to create the dataset more efficiently and report strong accuracy improvements of the trained models compared to image-level active learning baselines. The code is available at this https URL