Research map / Label-efficient and robust learning / Domain adaptation and generalization
Generalizing Gaze Estimation With Outlier-Guided Collaborative Adaptation
Yunfei Liu, Ruicong Liu, Haofei Wang, Feng Lu
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 |
|---|---|---|---|
| add PnP-GA to ResNet18 baseline | ResNet18 · ETH-XGaze to MPIIGaze | angular gaze error (deg): 8.767 -> 5.529 | Improved |
| add PnP-GA to ResNet18 baseline | ResNet18 · Gaze360 to EyeDiap | angular gaze error (deg): 8.977 -> 7.922 | Improved |
| add PnP-GA to SWCNN | SWCNN · ETH-XGaze to EyeDiap | angular gaze error (deg): 24.941 -> 15.794 | Improved |
| add PnP-GA to Dilated-Net | Dilated-Net · Gaze360 to MPIIGaze | angular gaze error (deg): 16.683 -> 15.461 | Improved |
| add outlier-guided model adaptation (oma) (ablation) | ResNet18 source only · ETH-XGaze to MPIIGaze | angular gaze error (deg): 8.77 -> 5.79 | Improved |
| add on-line adaptation (2oma) to oma (ablation) | CNN+oma · ETH-XGaze to MPIIGaze | angular gaze error (deg): 5.79 -> 5.65 | Improved |
| add Jensen-Shannon divergence (js) to oma (ablation) | CNN+oma · Gaze360 to MPIIGaze | angular gaze error (deg): 7.01 -> 6.62 | Improved |
| add source gaze constraint (sg) (ablation) | CNN+2oma+js · Gaze360 to MPIIGaze | angular gaze error (deg): 6.51 -> 6.18 | Improved |
| replace L1 loss with outlier-guided loss (ablation) | PnP-GA · ETH-XGaze to MPIIGaze | angular gaze error (deg): 7.00 -> 5.53 | Improved |
| replace L2 loss with outlier-guided loss (ablation) | PnP-GA · Gaze360 to EyeDiap | angular gaze error (deg): 8.23 -> 7.92 | Improved |
| use significance level 0.05 instead of 0.1 (ablation) | PnP-GA · ETH-XGaze to MPIIGaze | angular gaze error (deg): 6.11 -> 5.53 | Improved |
| use significance level 0.05 instead of 0.01 (ablation) | PnP-GA · Gaze360 to MPIIGaze | angular gaze error (deg): 6.97 -> 6.18 | Improved |
Closest papers by meaning
- PnP-GA+: Plug-and-Play Domain Adaptation for Gaze Estimation Using Model Variants (TPAMI 2024 · similarity 0.95)
- PureGaze: Purifying Gaze Feature for Generalizable Gaze Estimation (AAAI 2022 · similarity 0.90)
- Contrastive Regression for Domain Adaptation on Gaze Estimation (CVPR 2022 · similarity 0.89)
- Appearance-Based Gaze Estimation With Deep Learning: A Review and Benchmark (TPAMI 2024 · similarity 0.89)
- From Feature to Gaze: A Generalizable Replacement of Linear Layer for Gaze Estimation (CVPR 2024 · similarity 0.89)
- Gaze from Origin: Learning for Generalized Gaze Estimation by Embedding the Gaze Frontalization Process (AAAI 2024 · similarity 0.88)
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.
