Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2408.16964 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909301327527936 |
|---|---|
| author | Kim, Younghan Moon, Kangryun Park, Yongjun Kim, Yonggyu |
| author_facet | Kim, Younghan Moon, Kangryun Park, Yongjun Kim, Yonggyu |
| contents | The availability of extensive datasets containing gaze information for each subject has significantly enhanced gaze estimation accuracy. However, the discrepancy between domains severely affects a model's performance explicitly trained for a particular domain. In this paper, we propose the Causal Representation-Based Domain Generalization on Gaze Estimation (CauGE) framework designed based on the general principle of causal mechanisms, which is consistent with the domain difference. We employ an adversarial training manner and an additional penalizing term to extract domain-invariant features. After extracting features, we position the attention layer to make features sufficient for inferring the actual gaze. By leveraging these modules, CauGE ensures that the neural networks learn from representations that meet the causal mechanisms' general principles. By this, CauGE generalizes across domains by extracting domain-invariant features, and spurious correlations cannot influence the model. Our method achieves state-of-the-art performance in the domain generalization on gaze estimation benchmark. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_16964 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Causal Representation-Based Domain Generalization on Gaze Estimation Kim, Younghan Moon, Kangryun Park, Yongjun Kim, Yonggyu Computer Vision and Pattern Recognition The availability of extensive datasets containing gaze information for each subject has significantly enhanced gaze estimation accuracy. However, the discrepancy between domains severely affects a model's performance explicitly trained for a particular domain. In this paper, we propose the Causal Representation-Based Domain Generalization on Gaze Estimation (CauGE) framework designed based on the general principle of causal mechanisms, which is consistent with the domain difference. We employ an adversarial training manner and an additional penalizing term to extract domain-invariant features. After extracting features, we position the attention layer to make features sufficient for inferring the actual gaze. By leveraging these modules, CauGE ensures that the neural networks learn from representations that meet the causal mechanisms' general principles. By this, CauGE generalizes across domains by extracting domain-invariant features, and spurious correlations cannot influence the model. Our method achieves state-of-the-art performance in the domain generalization on gaze estimation benchmark. |
| title | Causal Representation-Based Domain Generalization on Gaze Estimation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2408.16964 |