Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
What a MESS: Multi-Domain Evaluation of Zero-Shot Semantic Segmentation
Benedikt Blumenstiel, Johannes Jakubik, Hilde Kuehne, Michael Vössing
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 |
|---|---|---|---|
| Use CAT-Seg-L instead of Grounded-SAM-L with Grounding DINO prompts | MESS multi-domain evaluation · MESS mean | mIoU: 29.05 -> 38.14 | Improved |
| Use RGB rather than thermal imagery | CAT-Seg-L on PST900 · PST900 | mIoU: 25.26 -> 65.55 | Improved |
| Map thermal data to a pseudo color scale instead of using thermal data directly | CAT-Seg-L on PST900 · PST900 | mIoU: 25.26 -> 25.43 | No clear effect |
| Use generic class terminology instead of domain-specific terminology | zero-shot segmentation models on Kvasir-Instrument · Kvasir-Instrument | mIoU | Improved |
| Merge similar classes into a single class | zero-shot segmentation models on datasets with similar labels · MESS datasets with similar classes | model performance | Improved |
| Use oracle bounding-box prompts instead of text-generated Grounding DINO prompts | SAM · MESS mean | mIoU: 29.05 -> 75.67 | Improved |
Closest papers by meaning
- DiffCut: Catalyzing Zero-Shot Semantic Segmentation with Diffusion Features and Recursive Normalized Cut (NeurIPS 2024 · similarity 0.90)
- Cascade-CLIP: Cascaded Vision-Language Embeddings Alignment for Zero-Shot Semantic Segmentation (ICML 2024 · similarity 0.89)
- Exploring Regional Clues in CLIP for Zero-Shot Semantic Segmentation (CVPR 2024 · similarity 0.88)
- Unlocking the Potential of Pre-trained Vision Transformers for Few-Shot Semantic Segmentation through Relationship Descriptors (CVPR 2024 · similarity 0.88)
- APSeg: Auto-Prompt Network for Cross-Domain Few-Shot Semantic Segmentation (CVPR 2024 · similarity 0.88)
- Label-Efficient Few-Shot Semantic Segmentation with Unsupervised Meta-Training (AAAI 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.
