Tackling Background Misclassification in Box-Supervised Segmentation: A Background Constraint Approach.
Zhicheng ZhangTakio KuritaPublished in: IW-FCV (2024)
Keyphrases
- foreground and background
- background pixels
- global thresholding
- segmentation algorithm
- complex background
- segmentation method
- foreground objects
- low depth of field
- foreground segmentation
- bounding box
- region growing
- computer vision
- learning algorithm
- level set
- fully unsupervised
- thresholding method
- image segmentation
- object regions
- moving object segmentation
- machine learning