Research map / Label-efficient and robust learning / Semi-supervised learning
XNet: Wavelet-Based Low and High Frequency Fusion Networks for Fully- and Semi-Supervised Semantic Segmentation of Biomedical Images
Yanfeng Zhou, Jiaxing Huang, Chenlong Wang, Le Song, Ge Yang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| feed wavelet LF and HF images through separate encoders with a fusion module (ablation) | XNet · GlaS fully-supervised | Jaccard index: 75.82 -> 84.77 | Improved |
| add LF/HF fusion module (semi-supervised) (ablation) | XNet · GlaS semi-supervised | Jaccard index: 66.78 -> 80.89 | Improved |
| use only HF images as input (ablation) | XNet · GlaS fully-supervised | Jaccard index: 82.03 -> 54.50 | Hurt |
| use LF and HF images as input without fusion (ablation) | XNet · GlaS semi-supervised | Jaccard index: 78.52 -> 66.78 | Hurt |
| use Db 2 wavelet basis instead of Dmey (ablation) | XNet · GlaS semi-supervised | Jaccard index: 73.95 -> 80.89 | Improved |
| use 2 LF and 2 HF fusion features rather than 3 and 3 (ablation) | XNet · GlaS semi-supervised | Jaccard index: 78.39 -> 80.89 | Improved |
| scale UNet up to XNet's parameter count (ablation) | UNet · GlaS fully-supervised | Jaccard index: 81.54 -> 82.73 | Improved |
| halve XNet channels (ablation) | XNet · GlaS fully-supervised | Jaccard index: 84.77 -> 83.49 | No clear effect |
| add unlabeled images to 20% labeled data (ablation) | XNet · GlaS | Jaccard index: 69.84 -> 80.89 | Improved |
| apply XNet to images with little high-frequency information (ISIC-2017) | XNet · ISIC-2017 | Jaccard index: 74.49 -> 73.94 | Hurt |
Later papers that compared with this method
1 later accepted paper reports XNet as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Semi-supervised Segmentation of Histopathology Images with Noise-Aware Topological Consistency (ECCV 2024 · better on 8 of 8 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- nnWNet: Rethinking the Use of Transformers in Biomedical Image Segmentation and Calling for a Unified Evaluation Benchmark (CVPR 2025 · similarity 0.87)
- FAMNet: Frequency-aware Matching Network for Cross-domain Few-shot Medical Image Segmentation (AAAI 2025 · similarity 0.85)
- SGTC: Semantic-Guided Triplet Co-training for Sparsely Annotated Semi-Supervised Medical Image Segmentation (AAAI 2025 · similarity 0.85)
- Steady Progress Beats Stagnation: Mutual Aid of Foundation and Conventional Models in Mixed Domain Semi-Supervised Medical Image Segmentation (CVPR 2025 · similarity 0.85)
- DuSSS: Dual Semantic Similarity-Supervised Vision-Language Model for Semi-Supervised Medical Image Segmentation (AAAI 2025 · similarity 0.85)
- Combinatorial CNN-Transformer Learning with Manifold Constraints for Semi-supervised Medical Image Segmentation (AAAI 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
