Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2506.05607 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915330489581568 |
|---|---|
| author | Lin, Shuchen Feng, Mingtao Dong, Weisheng Wu, Fangfang Luo, Jianqiao Wang, Yaonan Shi, Guangming |
| author_facet | Lin, Shuchen Feng, Mingtao Dong, Weisheng Wu, Fangfang Luo, Jianqiao Wang, Yaonan Shi, Guangming |
| contents | Real-world image super-resolution (Real-SR) is a challenging problem due to the complex degradation patterns in low-resolution images. Unlike approaches that assume a broadly encompassing degradation space, we focus specifically on achieving an optimal balance in how SR networks handle different degradation patterns within a fixed degradation space. We propose an improved paradigm that frames Real-SR as a data-heterogeneous multi-task learning problem, our work addresses task imbalance in the paradigm through coordinated advancements in task definition, imbalance quantification, and adaptive data rebalancing. Specifically, we introduce a novel task definition framework that segments the degradation space by setting parameter-specific boundaries for degradation operators, effectively reducing the task quantity while maintaining task discrimination. We then develop a focal loss based multi-task weighting mechanism that precisely quantifies task imbalance dynamics during model training. Furthermore, to prevent sporadic outlier samples from dominating the gradient optimization of the shared multi-task SR model, we strategically convert the quantified task imbalance into controlled data rebalancing through deliberate regulation of task-specific training volumes. Extensive quantitative and qualitative experiments demonstrate that our method achieves consistent superiority across all degradation tasks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_05607 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Controlled Data Rebalancing in Multi-Task Learning for Real-World Image Super-Resolution Lin, Shuchen Feng, Mingtao Dong, Weisheng Wu, Fangfang Luo, Jianqiao Wang, Yaonan Shi, Guangming Computer Vision and Pattern Recognition Real-world image super-resolution (Real-SR) is a challenging problem due to the complex degradation patterns in low-resolution images. Unlike approaches that assume a broadly encompassing degradation space, we focus specifically on achieving an optimal balance in how SR networks handle different degradation patterns within a fixed degradation space. We propose an improved paradigm that frames Real-SR as a data-heterogeneous multi-task learning problem, our work addresses task imbalance in the paradigm through coordinated advancements in task definition, imbalance quantification, and adaptive data rebalancing. Specifically, we introduce a novel task definition framework that segments the degradation space by setting parameter-specific boundaries for degradation operators, effectively reducing the task quantity while maintaining task discrimination. We then develop a focal loss based multi-task weighting mechanism that precisely quantifies task imbalance dynamics during model training. Furthermore, to prevent sporadic outlier samples from dominating the gradient optimization of the shared multi-task SR model, we strategically convert the quantified task imbalance into controlled data rebalancing through deliberate regulation of task-specific training volumes. Extensive quantitative and qualitative experiments demonstrate that our method achieves consistent superiority across all degradation tasks. |
| title | Controlled Data Rebalancing in Multi-Task Learning for Real-World Image Super-Resolution |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2506.05607 |