Research map / Label-efficient and robust learning / Semi-supervised learning
SemiCVT: Semi-Supervised Convolutional Vision Transformer for Semantic Segmentation
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Add SemiCVT's CNN-Transformer architecture and training scheme | ResNet-101 semi-supervised segmentation · PASCAL VOC classic, all labeled-data partitions | mIoU: +16.84%, +12.33%, +11.13%, +9.03% and +9.36% | Helped |
| Add SemiCVT's CNN-Transformer architecture and training scheme | ResNet-101 semi-supervised segmentation · PASCAL VOC blender, all partitions | mIoU: +0.99%, +0.94%, +0.90%, and +0.42% | Helped |
| Add SemiCVT's CNN-Transformer architecture and training scheme | ResNet-101 semi-supervised segmentation · Cityscapes, all partitions | mIoU: 1.89%, 1.04%, 0.70% and 0.50% | Helped |
| Add inter-model class-wise consistency (ablation) | SemiCVT · PASCAL VOC classic 1/4 and 1/2 | mIoU: 74.02 -> 74.99; 77.23 -> 78.54 | Helped |
| Add LGI Fourier-domain interactions in both directions (ablation) | CVT · PASCAL VOC classic 1/4 and 1/2 | mIoU: 68.38%; 73.68% | Helped |
| Replace LGI with Conformer spatial interaction (ablation) | CVT · PASCAL VOC classic 1/4 and 1/2 | mIoU: 65.88 -> 66.35; 71.69 -> 72.37; LGI: 68.38, 73.68 | Hurt |
| Replace LGI with MixFormer spatial interaction (ablation) | CVT · PASCAL VOC classic 1/4 and 1/2 | mIoU: 65.88 -> 66.63; 71.69 -> 72.56; LGI: 68.38, 73.68 | Hurt |
| Add the enhanced class-patch graph to class-wise consistency (ablation) | SemiCVT · PASCAL VOC classic 1/4 and 1/2 | mIoU: 74.02 -> 74.63; 77.23 -> 78.04 | Helped |
| Add the enhanced class-class graph to class-wise consistency (ablation) | SemiCVT · PASCAL VOC classic 1/4 and 1/2 | mIoU: 74.02 -> 74.46; 77.23 -> 77.91 | Helped |
| Combine enhanced class-patch and class-class graph prototypes (ablation) | SemiCVT · PASCAL VOC classic 1/4 and 1/2 | mIoU: +1.31% and +0.97% | Helped |
Closest papers by meaning
- Semi-supervised Vision Transformers at Scale (NeurIPS 2022 · similarity 0.91)
- Combinatorial CNN-Transformer Learning with Manifold Constraints for Semi-supervised Medical Image Segmentation (AAAI 2024 · similarity 0.89)
- Transferable Semi-Supervised Semantic Segmentation (AAAI 2018 · similarity 0.88)
- AllSpark: Reborn Labeled Features from Unlabeled in Transformer for Semi-Supervised Semantic Segmentation (CVPR 2024 · similarity 0.88)
- SemiDAViL: Semi-supervised Domain Adaptation with Vision-Language Guidance for Semantic Segmentation (CVPR 2025 · similarity 0.88)
- Semantic Segmentation With Generative Models: Semi-Supervised Learning and Strong Out-of-Domain Generalization (CVPR 2021 · similarity 0.88)
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.