Efficient Cost-and-Quality Controllable Arbitrary-scale Super-resolution with Fourier Constraints
This addresses a specific bottleneck in super-resolution for applications requiring controllable cost and quality, but it appears incremental as it builds on existing Fourier-based methods.
The paper tackled the problem of performance degradation and inefficiency in arbitrary-scale super-resolution by proposing a method that predicts multiple Fourier components jointly, improving both quality and efficiency compared to existing recurrent neural network approaches.
Cost-and-Quality (CQ) controllability in arbitrary-scale super-resolution is crucial. Existing methods predict Fourier components one by one using a recurrent neural network. However, this approach leads to performance degradation and inefficiency due to independent prediction. This paper proposes predicting multiple components jointly to improve both quality and efficiency.