Efficient Polyp Segmentation via Active Learning Framework
Proposing an active learning framework for polyp segmentation to minimize annotation costs and enhance model performance.
Multi-Layer Dense Attention Decoder for Polyp Segmentation Study
Proposing a novel decoder architecture for polyp segmentation using Dense Attention Gate and hierarchical feature aggregation.
Compact and Efficient Polyp Segmentation Model via Knowledge Distillation with Attention Supervision and Symmetrical Guiding
A compact and efficient model for polyp segmentation that leverages knowledge distillation with attention supervision and symmetrical guiding to achieve competitive performance with state-of-the-art methods.
Feedback Attention Network-v2 (FANetv2): An Advanced Encoder-Decoder Architecture for Accurate Polyp Segmentation
FANetv2 is an advanced encoder-decoder network that accurately segments polyps from colonoscopy images by leveraging an iterative feedback attention mechanism and integrating essential information about the number and size of polyps through a text-guided approach.
Enhanced Polyp Segmentation with Multi-scale Alignment and Frequency Domain Integration
The proposed PSTNet model integrates frequency domain cues and employs feature alignment techniques to enhance polyp segmentation accuracy, outperforming state-of-the-art methods.
HiFiSeg: Enhancing Polyp Segmentation Accuracy Using a Global-Local Vision Transformer
This paper introduces HiFiSeg, a novel deep learning model for colon polyp segmentation that leverages a global-local vision transformer framework to enhance the capture of high-frequency information, leading to improved accuracy in polyp detection and segmentation, especially for small targets and challenging datasets.
