Research map / Label-efficient and robust learning / Domain adaptation and generalization
Correlational Gaussian Processes for Cross-Domain Visual Recognition
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 |
|---|---|---|---|
| Use both positive and negative co-occurrences in the PL relational model rather than positive co-occurrences only (ablation) | CGPC · E-Album face recognition | Rank-1 accuracy (%): 95.24 -> 96.83 | Helped |
| Use both positive and negative co-occurrences in the PLE relational model rather than positive co-occurrences only (ablation) | CGPC · E-Album face recognition | Rank-1 accuracy (%): 95.24 -> 96.83 | Helped |
| Add high-order PLE to PP+PE (ablation) | CGPC · E-Album face recognition | Rank-1 accuracy (%): 97.62 -> 98.41 | Helped |
| Add high-order PLE to PP+PE (ablation) | CGPC · G-Album face recognition | Rank-1 accuracy (%): 90.56 -> 92.09 | Helped |
| Use RMGPC instead of SMGPC for the people-domain classifier (ablation) | CGPC · E-Album face recognition | Rank-1 accuracy (%): 97.62 -> 98.41 | Helped |
| Use RMGPC instead of SMGPC for the people-domain classifier (ablation) | CGPC · VP face recognition | Rank-1 accuracy (%): 69.01 -> 70.92 | Helped |
| Use CGPC with cross-domain relations rather than Lin-CDL | Face recognition · G-Album | Rank-1 accuracy (%): 85.46 -> 92.09 | Helped |
| Combine LE and PLE rather than using LE alone (ablation) | CGPC · E-Album location recognition | Rank-1 accuracy (%): 98.71 -> 100.00 | Helped |
| Combine GS and PGS rather than using GS alone (ablation) | CGPC · VP gesture recognition | Rank-1 accuracy (%): 43.13 -> 46.96 | Helped |
| Combine GS and PGS rather than using GS alone (ablation) | CGPC · VP scene recognition | Rank-1 accuracy (%): 57.19 -> 60.38 | Helped |
| Use CGPC rather than HContext for top-3 presence prediction | SUN 09 · SUN 09 | correctness (%): 38% -> 41.4% | Helped |
Closest papers by meaning
- Joint Geometrical and Statistical Alignment for Visual Domain Adaptation (CVPR 2017 · similarity 0.88)
- Unsupervised Correlation Analysis (CVPR 2018 · similarity 0.88)
- Convolutional Gaussian Processes (NeurIPS 2017 · similarity 0.86)
- Spotting the Unseen: Reciprocal Consensus Network Guided by Visual Archetypes (AAAI 2024 · similarity 0.84)
- Universal Domain Adaptive Object Detection via Dual Probabilistic Alignment (AAAI 2025 · similarity 0.84)
- Single Domain Generalization for Few-Shot Counting via Universal Representation Matching (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.