Video Ads Content Structuring by Combining Scene Confidence Prediction and Tagging.
Tomoyuki SuzukiAntonio Tejero-de-PablosPublished in: CoRR (2021)
Keyphrases
- video sequences
- video footage
- multimedia
- online video
- video images
- dynamic scenes
- metadata
- targeted advertising
- video data
- video content
- visual data
- input video
- video frames
- video scene
- multimedia data
- prediction accuracy
- video segments
- motion features
- moving camera
- scene change detection
- image sequences
- video streams
- d scene
- live video
- lecture videos
- surveillance videos
- image frames
- single image
- single frame
- image mosaics
- stationary camera
- keywords
- web content
- space time
- multimedia content
- video surveillance
- key frames
- moving objects
- confidence estimates
- conformal prediction
- video clips
- video analysis
- action recognition
- dynamic textures
- object motion
- visual features
- user experience
- tv programs
- virtual view
- social tagging
- user generated
- sponsored search