FBR-CNN: A Feedback Recurrent Network for Video Saliency Detection.
Guanqun DingNevrez ImamogluAli CaglayanMasahiro MurakawaRyosuke NakamuraPublished in: MLSP (2021)
Keyphrases
- saliency detection
- recurrent networks
- visual saliency
- saliency map
- recurrent neural networks
- video sequences
- object detection
- salient regions
- video content
- salient points
- detection method
- video frames
- multimedia
- natural images
- video data
- object level
- biologically inspired
- video surveillance
- human visual system
- feed forward
- real time
- visual cues
- pixel level
- object detectors
- visual attention
- video retrieval
- bayesian framework
- human detection
- eye movements
- input image
- multi class
- object recognition
- computer vision