AI Paper Atlas

Research map / Detection and segmentation / CNN features

ClassFormer: Exploring Class-Aware Dependency with Transformer for Medical Image Segmentation

Huimin Huang, Shiao Xie, Lanfen Lin, Ruofeng Tong, Yen-Wei Chen, Hong Wang, Yuexiang Li, Yawen Huang, Yefeng Zheng

AAAI 2023 · Detection and segmentation › CNN features · approach: Context modeling for semantic segmentation · 7 citations (OpenAlex)

Read the paper

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 changedWhereMeasuredEffect
Add ClassFormer to ResUNet (ablation)ResUNet · SynapseDSC: 78.69 -> 84.40Improved
Add the intra-class dynamic transformer (ablation)ResUNet · SynapseDSC: 78.69 -> 81.86Improved
Add Lbce supervision to the intra-class transformer (ablation)ResUNet with intra-class transformer · SynapseDSC: 81.86 -> 82.20Improved
Add the inter-class transformer (ablation)ResUNet · SynapseDSC: 78.69 -> 80.46Improved
Add class tokens to inter-class reasoning (ablation)ResUNet with inter-class transformer · SynapseDSC: 80.46 -> 82.77Improved
Add LEU to inter-class reasoning with class tokens (ablation)ResUNet with inter-class transformer and class tokens · SynapseDSC: 82.77 -> 83.07Improved
Arrange intra-class and inter-class transformers sequentially rather than in parallel (ablation)ClassFormer · SynapseDSC: 83.70 -> 83.74Improved
Use intra-class before inter-class reasoning (ablation)Sequential ClassFormer · SynapseDSC: 83.74 -> 84.40Improved
Add ClassFormer to U-Net (ablation)U-Net · SynapseDSC: 76.85 -> 83.89Improved
Add ClassFormer to TransUNet (ablation)TransUNet · SynapseDSC: 77.48 -> 81.25Improved
Add ClassFormer to SwinUNet (ablation)SwinUNet · SynapseDSC: 79.12 -> 80.59Improved
Use multi-scale salient-key selection (ablation)Intra-class dynamic transformer · SynapseDSCImproved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy