DSFMA: deeply supervised fully convolutional neural networks based on multi-level aggregation for saliency detection.
Inam UllahMuwei JianSumaira HussainJie GuoLi LianHui YuKashif ShaheedYilong YinPublished in: Multim. Tools Appl. (2021)
Keyphrases
- saliency detection
- convolutional neural networks
- object detection
- salient region detection
- salient regions
- detection method
- saliency map
- visual saliency
- pixel level
- learning algorithm
- semi supervised
- object level
- salient points
- supervised learning
- salient objects
- background subtraction
- image representation
- vision system
- image classification
- low level
- feature selection