TokenMotion: Motion-Guided Vision Transformer for Video Camouflaged Object Detection Via Learnable Token Selection.
Zifan YuErfan Bank TavakoliMeida ChenSuya YouRaghuveer RaoSanjeev AgarwalFengbo RenPublished in: CoRR (2023)
Keyphrases
- object detection
- space time
- computer vision
- spatial and temporal
- temporal filtering
- input video
- motion features
- object motion
- real time
- key frames
- video sequences
- image sequences
- visual perception
- visual cues
- shot change detection
- successive frames
- dynamic scenes
- motion estimation
- video frames
- temporal continuity
- moving camera
- spatio temporal
- low frame rate
- temporal coherence
- visual data
- motion detection
- motion analysis
- background subtraction
- optical flow
- static images
- vision system
- video objects
- temporal consistency
- motion trajectories
- motion model
- video surveillance
- dynamic textures
- action detection
- visual input
- moving objects
- video data
- video content
- multimedia
- layered representation
- face detection
- surveillance videos
- video shots
- motion patterns
- object recognition
- camera motion
- video summarization
- object segmentation
- human motion
- fuzzy logic
- scene understanding
- multi class
- video analysis
- pedestrian detection
- fault diagnosis
- video streams
- video scene
- motion segmentation
- human detection
- visual features
- visual information
- object tracking
- human activities
- motion capture
- depth map
- action recognition
- video retrieval
- object categories