Foreground Objects Detection Using a Fully Convolutional Network With a Background Model Image and Multiple Original Images.
Jae-Yeul KimJong-Eun HaPublished in: IEEE Access (2020)
Keyphrases
- foreground objects
- background model
- template matching
- pixel level
- background subtraction
- foreground detection
- background modeling
- pixel wise
- object segmentation
- object detection
- video sequences
- segmentation algorithm
- pixel values
- input image
- moving objects
- image data
- image segmentation
- video surveillance
- superpixels
- multiscale
- motion detection
- single image
- bounding box
- object recognition
- image features
- dynamic scenes
- moving camera
- multiple images
- image retrieval
- image matching
- target object
- matching algorithm
- image quality
- spatial information
- computer vision
- graph cuts
- image fusion
- video data
- segmentation method
- test images
- video frames