Motion-Aware Video Paragraph Captioning via Exploring Object-Centered Internal Knowledge.
Yimin HuGuorui YuYuejie ZhangRui FengTao ZhangXuequan LuShang GaoPublished in: ICASSP (2023)
Keyphrases
- object centered
- space time
- spatial and temporal
- video sequences
- video data
- motion trajectories
- motion features
- temporal coherence
- key frames
- image sequences
- knowledge representation
- static images
- knowledge acquisition
- temporal consistency
- visual data
- camera motion
- video frames
- multimedia
- video content
- motion analysis
- successive frames
- hierarchical representation
- moving camera
- input video
- layered representation
- knowledge base
- temporal continuity
- camera movement
- object motion
- surveillance videos
- reference frame
- motion patterns
- motion detection
- dynamic scenes
- video surveillance
- human motion
- event detection
- motion estimation
- domain knowledge
- moving objects