Research map / Label-efficient and robust learning / Domain adaptation and generalization
Hybrid-Domain Adaptative Representation Learning for Gaze Estimation
Qida Tan, Hongyu Yang, Wenchao Du
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 UDA loss (LUDA) (ablation) | dual-branch baseline · Cross-domain Gaze360->EyeDiap | MAE: 12.56 -> 9.49 | Improved |
| add UDA loss (LUDA) (ablation) | dual-branch baseline · EyeDiap in-domain | MAE: 5.54 -> 5.42 | Improved |
| add head-pose encoder (EPose) (ablation) | dual-branch baseline · Cross-domain Gaze360->MPIIFace | MAE: 11.97 -> 9.69 | Improved |
| add sparse-graph fusion (SGF) (ablation) | dual-branch baseline · Cross-domain Gaze360->EyeDiap | MAE: 12.56 -> 11.08 | Improved |
| add SGF to LUDA + EPose (ablation) | HARL · Cross-domain Gaze360->MPIIFace | MAE: 8.08 -> 7.49 | Improved |
| add UDA loss to EPose + SGF (ablation) | HARL · Cross-domain Gaze360->EyeDiap | MAE: 8.65 -> 7.63 | Improved |
| add EPose to LUDA + SGF (ablation) | HARL · EyeDiap in-domain | MAE: 5.17 -> 5.02 | Improved |
| add per-dimension subgraph in SGF (ablation) | HARL · EyeDiap in-domain | MAE: 5.55 -> 5.13 | Improved |
| use top-1 instead of top-3 edges in subgraph (ablation) | HARL · EyeDiap in-domain | MAE: 5.22 -> 5.02 | Improved |
| use HARL instead of CLIP-Gaze | cross-domain gaze estimation · Cross-domain Gaze360->MPIIFace | MAE: 6.89 -> 7.49 | Hurt |
Closest papers by meaning
- Contrastive Regression for Domain Adaptation on Gaze Estimation (CVPR 2022 · similarity 0.90)
- Unsupervised Gaze Representation Learning from Multi-view Face Images (CVPR 2024 · similarity 0.89)
- Gaze from Origin: Learning for Generalized Gaze Estimation by Embedding the Gaze Frontalization Process (AAAI 2024 · similarity 0.89)
- Enhancing 3D Gaze Estimation in the Wild using Weak Supervision with Gaze Following Labels (CVPR 2025 · similarity 0.88)
- De^2Gaze: Deformable and Decoupled Representation Learning for 3D Gaze Estimation (CVPR 2025 · similarity 0.88)
- OmniGaze: Reward-inspired Generalizable Gaze Estimation in the Wild (NeurIPS 2025 · similarity 0.87)
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.
