Research map / Detection and segmentation / Open-set and few-shot detection
Delving Into Shape-Aware Zero-Shot 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 |
|---|---|---|---|
| Apply SAZS instead of LSeg | DRN-based zero-shot segmentation · PASCAL-5ᶦ | mIoU: 47.4 -> 55.5 | Helped |
| Apply SAZS instead of LSeg | DRN-based zero-shot segmentation · COCO-20ᶦ | mIoU: 23.4 -> 35.2 | Helped |
| Apply SAZS instead of LSeg | ViT-L-based zero-shot segmentation · PASCAL-5ᶦ | mIoU: 52.3 -> 59.4 | Helped |
| Apply SAZS instead of LSeg | ViT-L-based zero-shot segmentation · COCO-20ᶦ | mIoU: 27.2 -> 35.3 | Helped |
| Add L_shape boundary-detection constraint (ablation) | SAZS with ViT backbone · PASCAL-5ᶦ | mIoU: 58.4 -> 59.4 | Helped |
| Add L_shape boundary-detection constraint (ablation) | SAZS with ViT backbone · COCO-20ᶦ | mIoU: 34.7 -> 35.3 | Helped |
| Add L_shape boundary-detection constraint (ablation) | SAZS with DRN backbone · COCO-20ᶦ | mIoU: 35.2 -> 35.2 | No clear effect |
| Add spectral eigensegment fusion (ablation) | SAZS with ViT backbone · PASCAL-5ᶦ | mIoU: 52.4 -> 59.4 | Helped |
| Add spectral eigensegment fusion (ablation) | SAZS with ViT backbone · COCO-20ᶦ | mIoU: 29.1 -> 35.3 | Helped |
| Add spectral eigensegment fusion (ablation) | SAZS with DRN backbone · COCO-20ᶦ | mIoU: 26.6 -> 35.2 | Helped |
| Use Z_shape alone instead of neither affinity component (ablation) | SAZS fusion · PASCAL-5ᶦ cross-dataset | mIoU: 58.4 -> 58.6 | Helped |
| Use Z_sem alone instead of neither affinity component (ablation) | SAZS fusion · PASCAL-5ᶦ cross-dataset | mIoU: 58.4 -> 62.7 | Helped |
Closest papers by meaning
- DiffCut: Catalyzing Zero-Shot Semantic Segmentation with Diffusion Features and Recursive Normalized Cut (NeurIPS 2024 · similarity 0.89)
- Unlocking the Potential of Pre-trained Vision Transformers for Few-Shot Semantic Segmentation through Relationship Descriptors (CVPR 2024 · similarity 0.89)
- Bridge the Points: Graph-based Few-shot Segment Anything Semantically (NeurIPS 2024 · similarity 0.88)
- Spectral Prompt Tuning: Unveiling Unseen Classes for Zero-Shot Semantic Segmentation (AAAI 2024 · similarity 0.88)
- Distilling Spectral Graph for Object-Context Aware Open-Vocabulary Semantic Segmentation (CVPR 2025 · similarity 0.87)
- Exploring Regional Clues in CLIP for Zero-Shot Semantic Segmentation (CVPR 2024 · 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.