AnySR: Realizing Image Super-Resolution as Any-Scale, Any-Resource
This work addresses efficiency issues in SISR applications for users needing scalable image enhancement, but it is incremental as it builds on existing arbitrary-scale methods.
The paper tackles the problem of improving efficiency and scalability in single-image super-resolution (SISR) by introducing AnySR, which rebuilds existing arbitrary-scale methods into any-scale, any-resource implementations, reducing resource requirements for smaller scales without additional parameters and performing on par with existing methods.
In an effort to improve the efficiency and scalability of single-image super-resolution (SISR) applications, we introduce AnySR, to rebuild existing arbitrary-scale SR methods into any-scale, any-resource implementation. As a contrast to off-the-shelf methods that solve SR tasks across various scales with the same computing costs, our AnySR innovates in: 1) building arbitrary-scale tasks as any-resource implementation, reducing resource requirements for smaller scales without additional parameters; 2) enhancing any-scale performance in a feature-interweaving fashion, inserting scale pairs into features at regular intervals and ensuring correct feature/scale processing. The efficacy of our AnySR is fully demonstrated by rebuilding most existing arbitrary-scale SISR methods and validating on five popular SISR test datasets. The results show that our AnySR implements SISR tasks in a computing-more-efficient fashion, and performs on par with existing arbitrary-scale SISR methods. For the first time, we realize SISR tasks as not only any-scale in literature, but also as any-resource. Code is available at https://github.com/CrispyFeSo4/AnySR.