arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1708.02191cs.CVcs.AI
收藏

无标记视频中人脸识别的无监督领域自适应方法

Unsupervised Domain Adaptation for Face Recognition in Unlabeled Videos

NEC Labs America(NEC美国实验室);UC Merced(加州大学默塞德分校);Dalian University of Technology(大连理工大学);UC San Diego(加州大学圣迭戈分校)

Kihyuk Sohn, Sifei Liu, Guangyu Zhong, Xiang Yu, Ming-Hsuan Yang, Manmohan Chandraker

摘要:

尽管人脸识别技术取得了快速的进步,但基于静态图像的人脸识别和基于视频的人脸识别的性能之间仍然存在明显的差距,这是由于这两个领域之间的视觉质量存在巨大差异,以及难以管理各种大规模视频数据集。本文解决了这两个挑战,通过图像到视频的特征级领域自适应的方法,学习歧视性的视频帧表示。该框架利用大规模的未标记的视频数据,以减少不同领域之间的差距,同时从大规模标记的静止图像转移的歧视性知识。给定在图像域中预训练的人脸识别网络,通过(i)通过特征匹配将知识从网络提取到视频自适应网络,(ii)通过合成数据增强执行特征恢复,以及(iii)通过域对抗训练学习域不变特征来实现自适应。我们通过鉴别器引导的特征融合进一步提高性能,该融合提高了高质量帧,同时消除了因视频域特定因素而退化的帧。在YouTube Faces和IJB-A数据集上的实验表明,每个模块都有助于我们的特征级领域自适应框架,并大大提高了视频人脸识别的性能,以达到最先进的准确性。我们定性地证明了网络学习抑制视频中的各种伪影,如姿势,照明或遮挡,而无需对其进行明确的训练。

英文摘要:

Despite rapid advances in face recognition, there remains a clear gap between the performance of still image-based face recognition and video-based face recognition, due to the vast difference in visual quality between the domains and the difficulty of curating diverse large-scale video datasets. This paper addresses both of those challenges, through an image to video feature-level domain adaptation approach, to learn discriminative video frame representations. The framework utilizes large-scale unlabeled video data to reduce the gap between different domains while transferring discriminative knowledge from large-scale labeled still images. Given a face recognition network that is pretrained in the image domain, the adaptation is achieved by (i) distilling knowledge from the network to a video adaptation network through feature matching, (ii) performing feature restoration through synthetic data augmentation and (iii) learning a domain-invariant feature through a domain adversarial discriminator. We further improve performance through a discriminator-guided feature fusion that boosts high-quality frames while eliminating those degraded by video domain-specific factors. Experiments on the YouTube Faces and IJB-A datasets demonstrate that each module contributes to our feature-level domain adaptation framework and substantially improves video face recognition performance to achieve state-of-the-art accuracy. We demonstrate qualitatively that the network learns to suppress diverse artifacts in videos such as pose, illumination or occlusion without being explicitly trained for them.

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。

↑