谢雪梅
研究方向
行为识别及目标检测 (Human Action Recognition & Object Detection)
场景理解与视频分析 (Scene understanding & Video analysis)
深度学习特征表示 (Deep learning & representation)
主持项目
自然基金重点项目——面向弱小目标的机器学习与智能认知方法(61836008)
自然基金面上项目——面向大数据的预期形状导向压缩感知成像的重建方法研究
陕西省基金项目——智能视频监测及人员危害行为预警技术
工作经历及教育背景
2007 至今 西安电子科技大学 教授 博导
2005-2007 西安电子科技大学 副教授
2004.3-2004.7 香港大学 助理研究员(R.A.)
1998-2004 香港大学 博士
1991-1994 西安电子科技大学 硕士
主要研究方向
Action Recognition
- Multimodal action recognition
- Knowledge embedded two-person action recognition
Action recognition includes predicting not only abnormal events like stealing, murdering and other criminals, but also the trends. Thus even we can see though the camera what people are thinking.
项目：智能视频监测及人员危害行为预警技术
Human Pose Estimation
Human pose estimation solves the problem of localizing human body parts in images. It can be applied to many computer vision tasks such as action recognition, human tracking, and it is the underlying key technique for many commercial applications, such as Motion Sensing Gaming, Augmented or Mixed Reality et cetera.
Human-object Interaction
- Network
- Application
Human-Object Interactions (HOI) detection aims to localize a person, an object, as well as identify the interaction between the person and the object.
项目：面向智联网机器人的人物交互关系研究
Scene Understanding & Object Detection
Image/video Reconstruction
Research Activity