Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | https://arxiv.org/abs/2407.04245 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866911945632776192 |
|---|---|
| author | Ho, Ming-Yang Wu, Che-Ming Wu, Min-Sheng Tseng, Yufeng Jane |
| author_facet | Ho, Ming-Yang Wu, Che-Ming Wu, Min-Sheng Tseng, Yufeng Jane |
| contents | Recent advancements in ultra-high-resolution unpaired image-to-image translation have aimed to mitigate the constraints imposed by limited GPU memory through patch-wise inference. Nonetheless, existing methods often compromise between the reduction of noticeable tiling artifacts and the preservation of color and hue contrast, attributed to the reliance on global image- or patch-level statistics in the instance normalization layers. In this study, we introduce a Dense Normalization (DN) layer designed to estimate pixel-level statistical moments. This approach effectively diminishes tiling artifacts while concurrently preserving local color and hue contrasts. To address the computational demands of pixel-level estimation, we further propose an efficient interpolation algorithm. Moreover, we invent a parallelism strategy that enables the DN layer to operate in a single pass. Through extensive experiments, we demonstrate that our method surpasses all existing approaches in performance. Notably, our DN layer is hyperparameter-free and can be seamlessly integrated into most unpaired image-to-image translation frameworks without necessitating retraining. Overall, our work paves the way for future exploration in handling images of arbitrary resolutions within the realm of unpaired image-to-image translation. Code is available at: https://github.com/Kaminyou/Dense-Normalization. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_04245 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Every Pixel Has its Moments: Ultra-High-Resolution Unpaired Image-to-Image Translation via Dense Normalization Ho, Ming-Yang Wu, Che-Ming Wu, Min-Sheng Tseng, Yufeng Jane Computer Vision and Pattern Recognition Recent advancements in ultra-high-resolution unpaired image-to-image translation have aimed to mitigate the constraints imposed by limited GPU memory through patch-wise inference. Nonetheless, existing methods often compromise between the reduction of noticeable tiling artifacts and the preservation of color and hue contrast, attributed to the reliance on global image- or patch-level statistics in the instance normalization layers. In this study, we introduce a Dense Normalization (DN) layer designed to estimate pixel-level statistical moments. This approach effectively diminishes tiling artifacts while concurrently preserving local color and hue contrasts. To address the computational demands of pixel-level estimation, we further propose an efficient interpolation algorithm. Moreover, we invent a parallelism strategy that enables the DN layer to operate in a single pass. Through extensive experiments, we demonstrate that our method surpasses all existing approaches in performance. Notably, our DN layer is hyperparameter-free and can be seamlessly integrated into most unpaired image-to-image translation frameworks without necessitating retraining. Overall, our work paves the way for future exploration in handling images of arbitrary resolutions within the realm of unpaired image-to-image translation. Code is available at: https://github.com/Kaminyou/Dense-Normalization. |
| title | Every Pixel Has its Moments: Ultra-High-Resolution Unpaired Image-to-Image Translation via Dense Normalization |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2407.04245 |