Recurrent Graph Neural Networks for Video Instance Segmentation.
Emil BrissmanJoakim JohnanderMartin DanelljanMichael FelsbergPublished in: Int. J. Comput. Vis. (2023)
Keyphrases
- neural network
- recurrent neural networks
- video segmentation
- feed forward
- image segmentation
- video sequences
- segmentation method
- text detection
- graph representation
- pattern recognition
- object segmentation and tracking
- segmentation algorithm
- level set
- multiscale
- low grade gliomas
- video content
- medical images
- video data
- fuzzy logic
- foreground background segmentation
- temporal segmentation
- video objects
- weighted graph
- random walk
- graph model
- multimedia
- segmentation accuracy
- normalized cut
- video scene
- min cut max flow
- real time
- video frames
- video streams
- video clips
- structured data
- objects in video sequences
- energy function
- edge detection
- spiking neural networks
- genetic algorithm
- directed graph
- neural network model
- video analysis
- motion estimation
- region growing
- event detection
- background subtraction
- connected components
- fully unsupervised
- key frames
- segmented regions
- graph theory
- space time