Robust Video Hashing Based on Local Fluctuation Preserving for Tracking Deep Fake Videos.
Lv ChenDengpan YeYueyun ShangJiaqing HuangPublished in: ICASSP (2022)
Keyphrases
- dynamic background
- video frames
- foreground background segmentation
- video sequences
- moving camera
- input video
- video images
- video content
- video data
- video surveillance
- video database
- video dataset
- video analysis
- video copy detection
- surveillance videos
- video clips
- long video sequences
- successive frames
- real time
- background subtraction
- partial occlusion
- robust tracking
- key frames
- real time face tracking
- human actions
- object segmentation and tracking
- video scene
- natural language descriptions
- event recognition
- face detection and tracking
- low frame rate
- stationary camera
- online video
- spatiotemporal features
- moving objects
- video streams
- video representation
- content based copy detection
- video editing
- temporal coherence
- object tracking
- video retrieval
- particle filter
- semantic concept detection
- video search
- video shots
- foreground objects
- dynamic scenes
- recognition of human actions
- visual tracking
- space time
- human activities
- action recognition
- object tracking algorithm
- video event
- image sequences
- high definition
- multimedia
- video indexing
- mean shift
- youtube videos
- spatial and temporal
- news video
- event detection
- video stabilization
- lecture videos