Research map / Label-efficient and robust learning / Domain adaptation and generalization
Joint Geometrical and Statistical Alignment for Visual Domain Adaptation
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 |
|---|---|---|---|
| Jointly align source and target subspaces while minimizing distribution shift | JGSA · Office+Caltech, SURF (average) | accuracy: 46.38 -> 50.58 | Helped |
| Jointly align source and target subspaces while minimizing distribution shift | JGSA · Office+Caltech, Decaf_6 (average) | accuracy: 88.18 -> 89.98 | Helped |
| Use coupled source and target projections for joint alignment | JGSA · MNIST/USPS (average) | accuracy: 63.47 -> 74.30 | Helped |
| Jointly reduce geometric and statistical domain shift | JGSA · RGB-D action datasets (average) | accuracy: 75.55 -> 77.81 | Helped |
| Use JGSA linear rather than KEMA | Office-Caltech Decaf_7 · 8 dataset pairs | accuracy: 89.1% -> 90.18% | Helped |
| Use RBF kernel rather than the primal version (ablation) | JGSA · Office+Caltech, SURF (average) | accuracy: 50.04 -> 50.58 | Helped |
| Use linear kernel rather than the primal version (ablation) | JGSA · Office+Caltech, SURF (average) | accuracy: 50.04 -> 50.18 | Helped |
| Use a kernelized version rather than the primal version (ablation) | JGSA · Office+Caltech, Decaf_6 (average) | accuracy: 89.98; 89.76; 88.97 | No clear effect |
| Vary the target-label refinement iteration count (ablation) | JGSA · RGB-D action recognition | accuracy | No clear effect |
| Choose subspace dimension k in [20,180] (ablation) | JGSA · USPS→MNIST, W→A, and MSR→MAD | accuracy: k∈ [20,180] | Helped |
| Choose regularization β in [2^-15,0.5] (ablation) | JGSA · USPS→MNIST, W→A, and MSR→MAD | accuracy: β∈ [2^{-15},0.5] | Helped |
Closest papers by meaning
- Learning an Invariant Hilbert Space for Domain Adaptation (CVPR 2017 · similarity 0.89)
- Cross-Domain Visual Representations via Unsupervised Graph Alignment (AAAI 2019 · similarity 0.89)
- Joint Domain Alignment and Discriminative Feature Learning for Unsupervised Deep Domain Adaptation (AAAI 2019 · similarity 0.88)
- Unified Domain Generalization and Adaptation for Multi-View 3D Object Detection (NeurIPS 2024 · similarity 0.87)
- Universal Domain Adaptive Object Detection via Dual Probabilistic Alignment (AAAI 2025 · similarity 0.87)
- Source-Free and Image-Only Unsupervised Domain Adaptation for Category Level Object Pose Estimation (ICLR 2024 · similarity 0.86)
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.