Decomposing style, content, and motion for videos.
Yaosi HuDacheng YinYuwang WangZhenzhong ChenChong LuoPublished in: J. Vis. Commun. Image Represent. (2022)
Keyphrases
- video footage
- moving camera
- motion features
- key frames
- dynamic scenes
- user generated
- input video
- recognition of human actions
- video frames
- motion model
- successive frames
- periodic motion
- video sequences
- motion estimation
- moving objects
- video content
- human motion
- video database
- motion capture data
- online video
- static images
- temporal coherence
- instructional videos
- motion analysis
- image sequences
- video data
- space time
- camera motion
- multimedia
- video representation
- motion patterns
- computer vision
- recognizing human actions
- video stabilization
- optical flow
- temporal domain
- video surveillance
- motion capture
- multimedia content
- spatio temporal
- web content
- surveillance videos
- motion parameters
- spatial and temporal
- video signals
- global motion
- action recognition
- metadata
- neighboring frames
- motion segmentation
- video clips
- traffic scenes
- lecture videos
- adjacent frames
- motion detection
- multi view
- particle filter
- user generated content
- video scene
- video segments
- video objects
- motion trajectories