MANet: a Motion-Driven Attention Network for Detecting the Pulse from a Facial Video with Drastic Motions.
Xuenan LiuXuezhi YangZiyan MengYe WangJie ZhangAlexander WongPublished in: ICCVW (2021)
Keyphrases
- body motions
- video sequences
- object motion
- space time
- image sequences
- human motion
- facial motion
- peer to peer
- motion features
- dynamic scenes
- motion capture data
- motion model
- optical flow
- motion patterns
- moving objects
- camera motion
- camera movement
- motion estimation
- face recognition
- key frames
- image motion
- motion trajectories
- network connectivity
- motion analysis
- video streams
- motion recognition
- activity recognition
- wireless sensor networks
- network structure
- spatial and temporal
- moving camera
- motion capture
- mobile ad hoc networks
- temporal continuity
- rigid motion
- motion segmentation
- dynamic textures
- human actions
- routing protocol
- motion parameters
- multibody
- flow field
- computer vision
- multiple moving objects
- visual attention
- video frames
- video data
- sensor networks
- long video sequences
- wireless medium