Rethinking Encoder-Decoder Flow Through Shared Structures
Fuente:
arXiv
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866916581833965568 |
|---|---|
| author | Laboyrie, Frederik Yucel, Mehmet Kerim Saa-Garriga, Albert |
| author_facet | Laboyrie, Frederik Yucel, Mehmet Kerim Saa-Garriga, Albert |
| contents | Dense prediction tasks have enjoyed a growing complexity of encoder architectures, decoders, however, have remained largely the same. They rely on individual blocks decoding intermediate feature maps sequentially. We introduce banks, shared structures that are used by each decoding block to provide additional context in the decoding process. These structures, through applying them via resampling and feature fusion, improve performance on depth estimation for state-of-the-art transformer-based architectures on natural and synthetic images whilst training on large-scale datasets. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_14535 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Rethinking Encoder-Decoder Flow Through Shared Structures Laboyrie, Frederik Yucel, Mehmet Kerim Saa-Garriga, Albert Computer Vision and Pattern Recognition Machine Learning Dense prediction tasks have enjoyed a growing complexity of encoder architectures, decoders, however, have remained largely the same. They rely on individual blocks decoding intermediate feature maps sequentially. We introduce banks, shared structures that are used by each decoding block to provide additional context in the decoding process. These structures, through applying them via resampling and feature fusion, improve performance on depth estimation for state-of-the-art transformer-based architectures on natural and synthetic images whilst training on large-scale datasets. |
| title | Rethinking Encoder-Decoder Flow Through Shared Structures |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2501.14535 |