第14届IEEE 国际人脸与手势识别大会(IEEE FG 2019)于5月14日至18日在法国里尔举行。实验室基于人脸的遥测式心率估计工作 “Robust Remote Heart Rate Estimation from Face Utilizing Spatial-temporal Attention” (作者:Xuesong Niu, Xingyuan Zhao, Hu Han, Abhijit Das, Antitza Dantcheva, Shiguang Shan and Xilin Chen)获得 IEEE FG 2019最佳海报论文奖,这是实验室研究生首次获得IEEE FG的最佳论文/海报论文奖项。该文章针对人脸视频心率估计中,心率信号微弱与不稳定的挑战,提出了一种基于注意力机制的心率估计算法,以及基于上下采样的心率数据分布增广方法,在VIPL-HR和MMSE-HR等代表性数据集上取得了当前最优的结果。
图1:基于人脸的心率估计方法框架
图2:最佳海报论文奖证书
此外,实验室另一篇关于人脸补全与超分辨联合学习的工作“FCSR-GAN: End-to-end Learning for Joint Face Completion and Super-resolution”(作者:Jiancheng Cai, Hu Han, Shiguang Shan, and Xilin Chen)也入选IEEE FG Best Reviewed Papers。
图3:人脸补全与超分辨联合学习算法流程
[1] Xuesong Niu, Xingyuan Zhao, Hu Han, Abhijit Das, Antitza Dantcheva, Shiguang Shan, Xilin Chen,"Robust Remote Heart Rate Estimation from Face Utilizing Spatial-temporal Attention," 14th IEEE International Conference on Automatic Face and Gesture Recognition (FG 2019), pp. 1-8, Lille, France, May 14-18, 2019. (最佳海报论文奖)
[2] Jiancheng Cai, Hu Han, Shiguang Shan, and Xilin Chen, "FCSR-GAN: End-to-end Learning for Joint Face Completion and Super-resolution," 14th IEEE International Conference on Automatic Face and Gesture Recognition (FG 2019), pp. 1-8, Lille, France, May 14-18, 2019. (Oral, 入选Best Reviewed Papers)
附件: