期刊
2022 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, ICIP
卷 -, 期 -, 页码 741-745出版社
IEEE
DOI: 10.1109/ICIP46576.2022.9897793
关键词
Face-Recognition; Head-Pose; Multi-Task-Learning; Knowledge-Distillation
资金
- Ford Motor Company
This paper introduces a novel approach to improving face-recognition pose-invariance by using semantic-segmentation features. Experimental evaluations show that the proposed Seg-Distilled-ID network outperforms three state-of-the-art encoders in terms of accuracy while using significantly fewer parameters.
This paper demonstrates a novel approach to improve face-recognition pose-invariance using semantic-segmentation features. The proposed Seg-Distilled-ID network jointly learns identification and semantic-segmentation tasks, where the segmentation task is then distilled (MobileNet encoder). Performance is benchmarked against three state-of-the-art encoders on a publicly available data-set emphasizing head-pose variations. Experimental evaluations show the Seg-Distilled-ID network shows notable robustness benefits, achieving 99.9% test-accuracy in comparison to 81.6% on ResNet-101, 96.1% on VGG-19 and 96.3% on InceptionV3. This is achieved using approximately one-tenth of the top encoder's inference parameters. These results demonstrate distilling semantic-segmentation features can efficiently address face-recognition pose-invariance.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据