SSGVS: Semantic Scene Graph-to-Video Synthesis.
Yuren CongJinhui YiBodo RosenhahnMichael Ying YangPublished in: CVPR Workshops (2023)
Keyphrases
- video sequences
- dynamic scenes
- semantic context
- video images
- video scene
- moving camera
- semantic concepts
- video annotation
- video event
- d scene
- video data
- scene change detection
- object detection and tracking
- graph model
- video content
- virtual view
- input video
- three dimensional
- live video
- space time
- video frames
- crowded scenes
- scene understanding
- natural language
- moving objects
- motion features
- real time
- image frames
- surveillance videos
- stationary camera
- video analysis
- directed graph
- graph structure
- video streams
- sports video
- multimedia
- image mosaics
- image sequences
- video footage
- single image
- object motion
- weighted graph
- semantic similarity
- visual data
- low level features
- outdoor scenes
- video surveillance
- complex scenes
- scene classification
- high level
- dynamic textures
- scene analysis
- multi view
- semantic labels
- layered representation
- high resolution
- semantic web
- motion estimation
- input image
- object detection
- visual features
- single frame
- bit rate
- image regions
- event detection
- spatial and temporal