Pulling Target to Source: A New Perspective on Domain Adaptive Semantic Segmentation
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2023
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866913560266801152 |
|---|---|
| author | Wang, Haochen Shen, Yujun Fei, Jingjing Li, Wei Wu, Liwei Wang, Yuxi Zhang, Zhaoxiang |
| author_facet | Wang, Haochen Shen, Yujun Fei, Jingjing Li, Wei Wu, Liwei Wang, Yuxi Zhang, Zhaoxiang |
| contents | Domain adaptive semantic segmentation aims to transfer knowledge from a labeled source domain to an unlabeled target domain. However, existing methods primarily focus on directly learning qualified target features, making it challenging to guarantee their discrimination in the absence of target labels. This work provides a new perspective. We observe that the features learned with source data manage to keep categorically discriminative during training, thereby enabling us to implicitly learn adequate target representations by simply \textbf{pulling target features close to source features for each category}. To this end, we propose T2S-DA, which we interpret as a form of pulling Target to Source for Domain Adaptation, encouraging the model in learning similar cross-domain features. Also, considering the pixel categories are heavily imbalanced for segmentation datasets, we come up with a dynamic re-weighting strategy to help the model concentrate on those underperforming classes. Extensive experiments confirm that T2S-DA learns a more discriminative and generalizable representation, significantly surpassing the state-of-the-art. We further show that our method is quite qualified for the domain generalization task, verifying its domain-invariant property. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2305_13752 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Pulling Target to Source: A New Perspective on Domain Adaptive Semantic Segmentation Wang, Haochen Shen, Yujun Fei, Jingjing Li, Wei Wu, Liwei Wang, Yuxi Zhang, Zhaoxiang Computer Vision and Pattern Recognition Domain adaptive semantic segmentation aims to transfer knowledge from a labeled source domain to an unlabeled target domain. However, existing methods primarily focus on directly learning qualified target features, making it challenging to guarantee their discrimination in the absence of target labels. This work provides a new perspective. We observe that the features learned with source data manage to keep categorically discriminative during training, thereby enabling us to implicitly learn adequate target representations by simply \textbf{pulling target features close to source features for each category}. To this end, we propose T2S-DA, which we interpret as a form of pulling Target to Source for Domain Adaptation, encouraging the model in learning similar cross-domain features. Also, considering the pixel categories are heavily imbalanced for segmentation datasets, we come up with a dynamic re-weighting strategy to help the model concentrate on those underperforming classes. Extensive experiments confirm that T2S-DA learns a more discriminative and generalizable representation, significantly surpassing the state-of-the-art. We further show that our method is quite qualified for the domain generalization task, verifying its domain-invariant property. |
| title | Pulling Target to Source: A New Perspective on Domain Adaptive Semantic Segmentation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2305.13752 |