Research map / Label-efficient and robust learning / Domain adaptation and generalization
Domain Generalization with MixStyle
Kaiyang Zhou, Yongxin Yang, Yu Qiao, Tao Xiang
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 |
|---|---|---|---|
| mix feature statistics across instances using MixStyle | ResNet-18 · PACS leave-one-domain-out | average accuracy: 79.5 -> 83.7 | Improved |
| mix feature statistics with random batch shuffling rather than domain labels | ResNet-18 · PACS leave-one-domain-out | average accuracy: 79.5 -> 82.8 | Improved |
| mix feature statistics with MixStyle | CNN · Digits-DG leave-one-domain-out | average accuracy: 73.7 -> 76.5 | Improved |
| mix feature statistics with MixStyle | ResNet-18 · Office-Home leave-one-domain-out | average accuracy: 64.7 -> 65.5 | Improved |
| add MixStyle | ResNet-50 · Market1501→Duke | mAP: 19.3 -> 23.8 | Improved |
| add MixStyle | ResNet-50 · Duke→Market1501 | mAP: 20.4 -> 24.7 | Improved |
| add MixStyle | OSNet · Market1501→Duke | mAP: 25.9 -> 27.3 | Improved |
| add MixStyle | OSNet · Duke→Market1501 | mAP: 24.0 -> 29.0 | Improved |
| apply MixStyle to residual blocks 1, 2, and 3 rather than only block 1 (ablation) | ResNet-18 · PACS category classification | accuracy: 80.1 -> 82.8 | Improved |
| also apply MixStyle after residual block 4 (ablation) | ResNet-18 · PACS category classification | accuracy: 82.8 -> 75.6 | Hurt |
| apply MixStyle at multiple lower-level blocks rather than blocks 1 and 4 (ablation) | ResNet-18 · PACS category classification | accuracy: 76.3 -> 81.7 | Improved |
| apply MixStyle after blocks 1 and 2 rather than only block 1 (ablation) | ResNet-50 · Cross-dataset person re-ID | mAP: 22.6 -> 23.8 | Improved |
Closest papers by meaning
- MixStyle Neural Networks for Domain Generalization and Adaptation (IJCV 2023 · similarity 0.96)
- Style Neophile: Constantly Seeking Novel Styles for Domain Generalization (CVPR 2022 · similarity 0.88)
- Learning To Diversify for Single Domain Generalization (ICCV 2021 · similarity 0.87)
- Mixup-Induced Domain Extrapolation for Domain Generalization (AAAI 2024 · similarity 0.85)
- On Unsupervised Domain Adaptation: Pseudo Label Guided Mixup for Adversarial Prompt Tuning (AAAI 2024 · similarity 0.85)
- DGFamba: Learning Flow Factorized State Space for Visual Domain Generalization (AAAI 2025 · 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.
