Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Generalizing Eye Tracking With Bayesian Adversarial Learning
Kang Wang, Rui Zhao, Hui Su, Qiang Ji
Check 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 an appearance-adversarial classifier (ablation) | CNN gaze estimator · Cross-subject on MPIIGaze, UT, Columbia, and EyeDiap | gaze estimation error: 12.2% | Improved |
| add a head-pose adversarial classifier alongside the appearance classifier (ablation) | CNN gaze estimator · Cross-pose on MPIIGaze, UT, Columbia, and EyeDiap | gaze estimation error: 14.4% | Improved |
| use Bayesian inference over multiple parameter samples instead of point estimation (ablation) | Adversarial CNN gaze estimator · Cross-subject on MPIIGaze, UT, Columbia, and EyeDiap | gaze estimation error: 21.9% | Improved |
| use Bayesian inference over multiple parameter samples instead of point estimation (ablation) | Adversarial CNN gaze estimator · Cross-pose on MPIIGaze, UT, Columbia, and EyeDiap | gaze estimation error: 21.1% | Improved |
| use Bayesian inference over multiple parameter samples instead of point estimation (ablation) | Adversarial CNN gaze estimator · Cross-dataset with UT as source | gaze estimation error: 17.9% | Improved |
| add a head-pose adversarial classifier (ablation) | CNN gaze estimator with appearance classifier · Cross-subject evaluation | gaze estimation error: 12.2% -> 15.6% | Improved |
| add a head-pose adversarial classifier (ablation) | CNN gaze estimator with appearance classifier · Cross-dataset evaluation | gaze estimation error: 10.1% -> 12.4% | Improved |
| increase the percentage of labeled target-domain data (ablation) | Proposed method · UT cross-subject and cross-pose | gaze estimation error | Improved |
| use the proposed adaptation method instead of fine-tuning (ablation) | UT target-domain adaptation · UT cross-subject and cross-pose | gaze estimation error | Improved |
Closest papers by meaning
- Learning a Generalized Gaze Estimator from Gaze-Consistent Feature (AAAI 2023 · similarity 0.89)
- From Feature to Gaze: A Generalizable Replacement of Linear Layer for Gaze Estimation (CVPR 2024 · similarity 0.88)
- Generalizing Gaze Estimation With Outlier-Guided Collaborative Adaptation (ICCV 2021 · similarity 0.88)
- PureGaze: Purifying Gaze Feature for Generalizable Gaze Estimation (AAAI 2022 · similarity 0.88)
- Gaze from Origin: Learning for Generalized Gaze Estimation by Embedding the Gaze Frontalization Process (AAAI 2024 · similarity 0.88)
- Enhancing 3D Gaze Estimation in the Wild using Weak Supervision with Gaze Following Labels (CVPR 2025 · 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.
