Clustering based Contrastive Learning for Improving Face Representations

Clustering based Contrastive Learning for Improving Face Representations
复制标题

DOI:
10.1109/fg47880.2020.00011
复制
发表时间:
2020-04
期刊:
2020 15th IEEE International Conference on Automatic Face and Gesture Recognition (FG 2020)
影响因子:
--
通讯作者:
Vivek Sharma;Makarand Tapaswi;M. Sarfraz;R. Stiefelhagen
Vivek Sharma;Makarand Tapaswi;M. Sarfraz;R. Stiefelhagen
中科院分区:
其他
文献类型:
--
作者:
Vivek Sharma;Makarand Tapaswi;M. Sarfraz;R. Stiefelhagen

文献摘要

相似文献

好的聚类算法可以发现数据中的自然分组。如果明智地使用这些分组,将为学习表示提供一种弱监督形式。在这项工作中,我们提出了基于聚类的对比学习(CCL),这是一种新的基于聚类的表示学习方法,它使用从聚类获得的标签以及视频约束来学习有区别的面部特征。我们在学习视频人脸聚类表示的挑战性任务中展示了我们的方法。通过多项消融研究,我们分析了从不同来源创建成对的正面和负面标签的影响。在三个具有挑战性的视频人脸聚类数据集(BBT-0101、BF-0502 和 ACCIO)上进行的实验表明,CCL 在所有数据集上都实现了新的最先进水平。
A good clustering algorithm can discover natural groupings in data. These groupings, if used wisely, provide a form of weak supervision for learning representations. In this work, we present Clustering-based Contrastive Learning (CCL), a new clustering-based representation learning approach that uses labels obtained from clustering along with video constraints to learn discriminative face features. We demonstrate our method on the challenging task of learning representations for video face clustering. Through several ablation studies, we analyze the impact of creating pair-wise positive and negative labels from different sources. Experiments on three challenging video face clustering datasets: BBT-0101, BF-0502, and ACCIO show that CCL achieves a new state-of-the-art on all datasets.