Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2503.12764 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915359807766528 |
|---|---|
| author | Liu, Yidi Li, Dong Ma, Yuxin Huang, Jie Zhang, Wenlong Fu, Xueyang Zha, Zheng-jun |
| author_facet | Liu, Yidi Li, Dong Ma, Yuxin Huang, Jie Zhang, Wenlong Fu, Xueyang Zha, Zheng-jun |
| contents | Ultra-high-definition (UHD) image restoration often faces computational bottlenecks and information loss due to its extremely high resolution. Existing studies based on Variational Autoencoders (VAE) improve efficiency by transferring the image restoration process from pixel space to latent space. However, degraded components are inherently coupled with background elements in degraded images, both information loss during compression and information gain during compensation remain uncontrollable. These lead to restored images often exhibiting image detail loss and incomplete degradation removal. To address this issue, we propose a Controlled Differential Disentangled VAE, which utilizes Hierarchical Contrastive Disentanglement Learning and an Orthogonal Gated Projection Module to guide the VAE to actively discard easily recoverable background information while encoding more difficult-to-recover degraded information into the latent space. Additionally, we design a Complex Invertible Multiscale Fusion Network to handle background features, ensuring their consistency, and utilize a latent space restoration network to transform the degraded latent features, leading to more accurate restoration results. Extensive experimental results demonstrate that our method effectively alleviates the information loss problem in VAE models while ensuring computational efficiency, significantly improving the quality of UHD image restoration, and achieves state-of-the-art results in six UHD restoration tasks with only 1M parameters. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_12764 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Decouple to Reconstruct: High Quality UHD Restoration via Active Feature Disentanglement and Reversible Fusion Liu, Yidi Li, Dong Ma, Yuxin Huang, Jie Zhang, Wenlong Fu, Xueyang Zha, Zheng-jun Computer Vision and Pattern Recognition Ultra-high-definition (UHD) image restoration often faces computational bottlenecks and information loss due to its extremely high resolution. Existing studies based on Variational Autoencoders (VAE) improve efficiency by transferring the image restoration process from pixel space to latent space. However, degraded components are inherently coupled with background elements in degraded images, both information loss during compression and information gain during compensation remain uncontrollable. These lead to restored images often exhibiting image detail loss and incomplete degradation removal. To address this issue, we propose a Controlled Differential Disentangled VAE, which utilizes Hierarchical Contrastive Disentanglement Learning and an Orthogonal Gated Projection Module to guide the VAE to actively discard easily recoverable background information while encoding more difficult-to-recover degraded information into the latent space. Additionally, we design a Complex Invertible Multiscale Fusion Network to handle background features, ensuring their consistency, and utilize a latent space restoration network to transform the degraded latent features, leading to more accurate restoration results. Extensive experimental results demonstrate that our method effectively alleviates the information loss problem in VAE models while ensuring computational efficiency, significantly improving the quality of UHD image restoration, and achieves state-of-the-art results in six UHD restoration tasks with only 1M parameters. |
| title | Decouple to Reconstruct: High Quality UHD Restoration via Active Feature Disentanglement and Reversible Fusion |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2503.12764 |