Research map / Point clouds / Segmentation and detection
Transformer-Based Visual Segmentation: A Survey
Xiangtai Li, Henghui Ding, Haobo Yuan, Wenwei Zhang, Jiangmiao Pang, Guangliang Cheng, Kai Chen, Ziwei Liu, Chen Change Loy
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 |
|---|---|---|---|
| add LSJ augmentation during training (ablation) | K-Net · COCO panoptic | PQ: 47.1 -> 47.5 | Improved |
| add LSJ augmentation during training (ablation) | K-Net · COCO instance | mAP: 39.1 -> 40.2 | Improved |
| add deformable FPN (ablation) | K-Net · COCO panoptic | PQ: 47.1 -> 47.9 | Improved |
| add deformable FPN (ablation) | K-Net · COCO instance | mAP: 39.1 -> 42.1 | Improved |
| add both LSJ and deformable FPN (ablation) | K-Net · COCO panoptic | PQ: 47.1 -> 48.5 | Improved |
| train 50 epochs instead of 36 with both techniques (K-Net++) (ablation) | K-Net · COCO panoptic | PQ: 48.5 -> 49.2 | Improved |
| add LSJ augmentation (ablation) | Mask2Former · ADE20k | mIoU: 47.0 -> 47.5 | Improved |
| add deformable FPN (removing it lowers score) (ablation) | Mask2Former · ADE20k | mIoU: 47.0 -> 44.2 | Improved |
| add LSJ augmentation (ablation) | Mask2Former · VSPW | mIoU: 39.6 -> 39.5 | No clear effect |
| add deformable FPN (removing it lowers score) (ablation) | Mask2Former · VSPW | mIoU: 39.6 -> 37.5 | Improved |
| OneFormer3D unified model vs SPFormer instance-only | OneFormer3D · ScanNet | AP50: 73.9 -> 78.1 | Improved |
| OneFormer3D vs SPFormer instance-only | OneFormer3D · S3DIS | AP50: 66.8 -> 72.0 | Improved |
Closest papers by meaning
- Segmenter: Transformer for Semantic Segmentation (ICCV 2021 · similarity 0.91)
- FeedFormer: Revisiting Transformer Decoder for Efficient Semantic Segmentation (AAAI 2023 · similarity 0.90)
- Rethinking Decoders for Transformer-based Semantic Segmentation: A Compression Perspective (NeurIPS 2024 · similarity 0.89)
- Learning Hierarchical Image Segmentation For Recognition and By Recognition (ICLR 2024 · similarity 0.88)
- Your ViT is Secretly an Image Segmentation Model (CVPR 2025 · similarity 0.88)
- Improving Transformer Based Line Segment Detection with Matched Predicting and Re-ranking (AAAI 2025 · similarity 0.87)
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.
