NGDNet: Nonuniform Gaussian-label distribution learning for infrared head pose estimation and on-task behavior understanding in the classroom

Neurocomputing(2021)

引用 83|浏览10
暂无评分
摘要
•A simple yet efficient nonuniform Gaussian-label distribution learning method is proposed the infrared HPE.•The proposed method reveals the essential properties from two different aspects: difference and nonuniform.•The robustness of the proposed model is verified by extensive experiments.
更多
查看译文
关键词
Head pose estimation,Deep learning,On-task behavior understanding,Infrared imaging,Soft label,Human–computer interaction,K-12 education
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要