Camera Lens Super-Resolution
This work provides a practical solution for improving super-resolution in real-world imaging systems like DSLRs and smartphones, though it is incremental as it builds on existing SR methods.
The paper tackles the problem of single image super-resolution (SR) by addressing the resolution-field-of-view tradeoff in realistic camera lenses, proposing CameraSR to reverse this degradation using real low- and high-resolution image pairs, and demonstrates its superiority over synthetic degradation models on the City100 dataset.
Existing methods for single image super-resolution (SR) are typically evaluated with synthetic degradation models such as bicubic or Gaussian downsampling. In this paper, we investigate SR from the perspective of camera lenses, named as CameraSR, which aims to alleviate the intrinsic tradeoff between resolution (R) and field-of-view (V) in realistic imaging systems. Specifically, we view the R-V degradation as a latent model in the SR process and learn to reverse it with realistic low- and high-resolution image pairs. To obtain the paired images, we propose two novel data acquisition strategies for two representative imaging systems (i.e., DSLR and smartphone cameras), respectively. Based on the obtained City100 dataset, we quantitatively analyze the performance of commonly-used synthetic degradation models, and demonstrate the superiority of CameraSR as a practical solution to boost the performance of existing SR methods. Moreover, CameraSR can be readily generalized to different content and devices, which serves as an advanced digital zoom tool in realistic imaging systems. Codes and datasets are available at https://github.com/ngchc/CameraSR.