PEM360: A dataset of 360° videos with continuous Physiological measurements, subjective Emotional ratings and Motion traces.
Florent RobertQuentin GuimardLucile SassatelliHui-Yin WuMarco WincklerAuriane GrosPublished in: IMX Workshops (2023)
Keyphrases
- human actions
- periodic motion
- space time
- recognition of human actions
- input video
- human motion
- motion features
- moving camera
- video sequences
- netflix prize
- motion capture data
- dynamic scenes
- action recognition in videos
- motion analysis
- motion model
- motion estimation
- static images
- successive frames
- camera motion
- action recognition
- collaborative filtering
- key frames
- measurement errors
- video stabilization
- optical flow
- motion segmentation
- motion patterns
- traffic scenes
- dynamic textures
- recognizing human actions
- spatio temporal
- motion cues
- motion capture
- image measurements
- multimedia event detection
- temporal domain
- video dataset
- global motion
- visual data
- video frames
- spatial and temporal
- emotional state
- motion trajectories
- image sequences
- image motion
- object motion
- video representation
- adjacent frames
- recommender systems
- video content
- video data
- video summarization
- heart rate
- moving objects
- computer vision
- activity recognition
- human activities
- event detection
- motion field
- temporal coherence
- photo collections
- event recognition