Robust Subjective Visual Property Prediction from Crowdsourced Pairwise Labels
Robust Subjective Visual Property Prediction from Crowdsourced Pairwise Labels
复制标题
根据众包成对标签进行稳健的主观视觉属性预测
DOI：
10.1109/tpami.2015.2456887
复制
发表时间：
2016-03-01
影响因子：
23.6
通讯作者：
Yao, Yuan
中科院分区：
文献类型：
--
作者：
Fu, Yanwei;Hospedales, Timothy M.;Yao, Yuan
从图像和视频中估计主观视觉特性的问题引起了越来越多的关注。主观视觉属性本身（例如图像和视频兴趣度）或作为视觉识别的中间表示（例如相对属性）是有用的。由于其模糊的性质，注释主观视觉属性的值以学习预测模型是具有挑战性的。为了使注释更可靠，最近的研究采用众包工具来收集成对比较标签。然而，使用众包数据也会引入离群值。现有方法依赖于多数投票来修剪注释离群值/错误。因此，它们需要收集大量的成对标签。更重要的是，作为一种局部离群点检测方法，多数投票在识别可能导致全局排名不一致的离群点方面是无效的。在本文中，我们提出了一种更有原则的方法来识别注释离群值，通过将主观视觉属性预测任务制定为统一的鲁棒学习排名问题，同时解决离群值检测和学习排名。这与现有方法的不同之处在于：（1）所提出的方法将局部成对比较标签集成在一起，以最小化对应于排名顺序的全局不一致性的成本，以及（2）离群点检测和学习排名问题联合解决。这不仅可以更好地检测注释离群值，还可以使用极其稀疏的注释进行学习。
The problem of estimating subjective visual properties from image and video has attracted increasing interest. A subjective visual property is useful either on its own (e.g. image and video interestingness) or as an intermediate representation for visual recognition (e.g. a relative attribute). Due to its ambiguous nature, annotating the value of a subjective visual property for learning a prediction model is challenging. To make the annotation more reliable, recent studies employ crowdsourcing tools to collect pairwise comparison labels. However, using crowdsourced data also introduces outliers. Existing methods rely on majority voting to prune the annotation outliers/errors. They thus require a large amount of pairwise labels to be collected. More importantly as a local outlier detection method, majority voting is ineffective in identifying outliers that can cause global ranking inconsistencies. In this paper, we propose a more principled way to identify annotation outliers by formulating the subjective visual property prediction task as a unified robust learning to rank problem, tackling both the outlier detection and learning to rank jointly. This differs from existing methods in that (1) the proposed method integrates local pairwise comparison labels together to minimise a cost that corresponds to global inconsistency of ranking order, and (2) the outlier detection and learning to rank problems are solved jointly. This not only leads to better detection of annotation outliers but also enables learning with extremely sparse annotations.