Kavli Affiliate: Yi Zhou
| First 5 Authors: Yunqi Gu, Tao Zhou, Yizhe Zhang, Yi Zhou, Kelei He
| Summary:
Automatic polyp segmentation plays a crucial role in the early diagnosis and
treatment of colorectal cancer (CRC). However, existing methods heavily rely on
fully supervised training, which requires a large amount of labeled data with
time-consuming pixel-wise annotations. Moreover, accurately segmenting polyps
poses challenges due to variations in shape, size, and location. To address
these issues, we propose a novel Dual-scale Enhanced and Cross-generative
consistency learning framework for semi-supervised polyp Segmentation (DEC-Seg)
from colonoscopy images. First, we propose a Cross-level Feature Aggregation
(CFA) module that integrates cross-level adjacent layers to enhance the feature
representation ability across different resolutions. To address scale
variation, we present a scale-enhanced consistency constraint, which ensures
consistency in the segmentation maps generated from the same input image at
different scales. This constraint helps handle variations in polyp sizes and
improves the robustness of the model. Additionally, we design a scale-aware
perturbation consistency scheme to enhance the robustness of the mean teacher
model. Furthermore, we propose a cross-generative consistency scheme, in which
the original and perturbed images can be reconstructed using cross-segmentation
maps. This consistency constraint allows us to mine effective feature
representations and boost the segmentation performance. To produce more
accurate segmentation maps, we propose a Dual-scale Complementary Fusion (DCF)
module that integrates features from two scale-specific decoders operating at
different scales. Extensive experimental results on five benchmark datasets
demonstrate the effectiveness of our DEC-Seg against other state-of-the-art
semi-supervised segmentation approaches. The implementation code will be
released at https://github.com/taozh2017/DECSeg.
| Search Query: ArXiv Query: search_query=au:”Yi Zhou”&id_list=&start=0&max_results=3