Multimodal Video-to-Near-Scene Annotation.
Chien-Li ChouHua-Tsung ChenSuh-Yin LeePublished in: IEEE Trans. Multim. (2017)
Keyphrases
- video sequences
- video images
- dynamic scenes
- video scene
- video annotation
- video data
- scene change detection
- multimedia
- input video
- visual data
- moving camera
- weakly labeled
- d scene
- object motion
- video material
- image mosaics
- video streams
- video content
- video frames
- moving objects
- three dimensional
- multi modal
- scene analysis
- live video
- multiple images
- motion features
- video footage
- active learning
- object detection and tracking
- image frames
- dynamic textures
- surveillance videos
- digital photos
- image sequences
- automatic annotation
- stationary camera
- metadata
- single image
- real time
- video analysis
- image annotation
- semantic annotation
- key frames
- audio visual
- story segmentation
- motion estimation
- video clips
- video retrieval
- object detection
- camera motion
- light source
- spatial and temporal
- semantic labels
- complex scenes
- crowded scenes
- photo collections
- multiple objects
- video database