Research map / Image generation / Image restoration
Joint Recovery of Dense Correspondence and Cosegmentation in Two Images
Check 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 hierarchical modeling and inference instead of the single-layer model (ablation) | Joint cosegmentation and correspondence method · FG3DCar | FAcc (5-pixel threshold): 0.727 -> 0.829 | Helped |
| add hierarchical modeling and inference instead of the single-layer model (ablation) | Joint cosegmentation and correspondence method · JODS | FAcc (5-pixel threshold): 0.473 -> 0.595 | Helped |
| add hierarchical modeling and inference instead of the single-layer model (ablation) | Joint cosegmentation and correspondence method · PASCAL | FAcc (5-pixel threshold): 0.414 -> 0.483 | Helped |
| add hierarchical modeling and inference instead of the single-layer model (ablation) | Joint cosegmentation and correspondence method · FG3DCar | SAcc (IoU): 0.757 -> 0.756 | No clear effect |
| add hierarchical modeling and inference instead of the single-layer model (ablation) | Joint cosegmentation and correspondence method · JODS | SAcc (IoU): 0.499 -> 0.504 | Helped |
| add hierarchical modeling and inference instead of the single-layer model (ablation) | Joint cosegmentation and correspondence method · PASCAL | SAcc (IoU): 0.616 -> 0.649 | Helped |
| jointly infer correspondence and cosegmentation rather than estimate correspondence alone | Joint cosegmentation and correspondence method · FG3DCar | FAcc (5-pixel threshold): 0.829 vs 0.634 | Helped |
| jointly infer correspondence and cosegmentation rather than estimate correspondence alone | Joint cosegmentation and correspondence method · JODS | FAcc (5-pixel threshold): 0.595 vs 0.522 | Helped |
| jointly infer correspondence and cosegmentation rather than use the strongest listed cosegmentation baseline | Joint cosegmentation and correspondence method · PASCAL | SAcc (IoU): 0.649 vs 0.500 | Helped |
Closest papers by meaning
- DenseReg: Fully Convolutional Dense Shape Regression In-The-Wild (CVPR 2017 · similarity 0.85)
- Semantically Coherent Co-Segmentation and Reconstruction of Dynamic Scenes (CVPR 2017 · similarity 0.85)
- DeMatch: Deep Decomposition of Motion Field for Two-View Correspondence Learning (CVPR 2024 · similarity 0.85)
- Adapting Dense Matching for Homography Estimation with Grid-based Acceleration (CVPR 2025 · similarity 0.84)
- FastJAM: a Fast Joint Alignment Model for Images (NeurIPS 2025 · similarity 0.84)
- SemAlign3D: Semantic Correspondence between RGB-Images through Aligning 3D Object-Class Representations (CVPR 2025 · similarity 0.84)
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.