Research map / Point clouds / Segmentation and detection
Weakly Supervised Scene Parsing with Point-Based Distance Metric Learning
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 cross-image PDML to PointSup (ablation) | PointSup with ResNet101 · PASCAL-Context validation | mIoU: 27.9 -> 29.7 | Helped |
| add cross-image PDML to PointSup (ablation) | PointSup with ResNet101 · ADE 20K validation | mIoU: 17.7 -> 19.0 | Helped |
| add online extension to PointSup + PDML (ablation) | PointSup + PDML with ResNet101 · PASCAL-Context validation | mIoU: 29.7 -> 30.0 | Helped |
| add online extension to PointSup + PDML (ablation) | PointSup + PDML with ResNet101 · ADE 20K validation | mIoU: 19.0 -> 19.6 | Helped |
| replace score-and-region extension with superpixel extension (ablation) | PDML with ResNet101 · PASCAL-Context validation | mIoU: 30.0 -> 26.7 | Hurt |
| replace score-and-region extension with K-Means extension (ablation) | PDML with ResNet101 · PASCAL-Context validation | mIoU: 30.0 -> 12.9 | Hurt |
| use score extension only at threshold 0.7, without region extension (ablation) | PDML with ResNet101 · PASCAL-Context validation | mIoU: 30.0 -> 16.7 | Hurt |
| use region extension only (ablation) | PDML with ResNet101 · PASCAL-Context validation | mIoU: 27.9 -> 29.7 | Helped |
| set score threshold to 0.8 with region extension instead of 0.7 (ablation) | PDML with ResNet101 · PASCAL-Context validation | mIoU: 30.0 -> 29.4 | Hurt |
| set score threshold to 0.9 with region extension instead of 0.7 (ablation) | PDML with ResNet101 · PASCAL-Context validation | mIoU: 30.0 -> 29.4 | Hurt |
| set score threshold to 0.6 with region extension instead of 0.7 (ablation) | PDML with ResNet101 · PASCAL-Context validation | mIoU: 30.0 -> 29.8 | Hurt |
| remove the positive-pair loss constraint (alpha 0.8 to 0.0) (ablation) | PDML with ResNet101 · PASCAL-Context validation | mIoU: 29.7 -> 26.2 | Hurt |
Closest papers by meaning
- Deep Structured Scene Parsing by Learning With Image Descriptions (CVPR 2016 · similarity 0.90)
- Weakly Supervised Visual Semantic Parsing (CVPR 2020 · similarity 0.88)
- Sample and Filter: Nonparametric Scene Parsing via Efficient Filtering (CVPR 2016 · similarity 0.86)
- Weakly Supervised Semantic Segmentation for Driving Scenes (AAAI 2024 · similarity 0.85)
- Semantic-aware SAM for Point-Prompted Instance Segmentation (CVPR 2024 · similarity 0.85)
- Curriculum Point Prompting for Weakly-Supervised Referring Image Segmentation (CVPR 2024 · similarity 0.85)
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.