Thera: Aliasing-Free Arbitrary-Scale Super-Resolution with Neural Heat Fields
This addresses aliasing issues in super-resolution for image processing applications, offering a novel solution with theoretical guarantees.
The paper tackles aliasing in arbitrary-scale super-resolution by introducing neural heat fields that model the point spread function exactly, enabling anti-aliasing at any resolution without extra cost, and Thera outperforms existing methods with better parameter efficiency.
Recent approaches to arbitrary-scale single image super-resolution (ASR) use neural fields to represent continuous signals that can be sampled at arbitrary resolutions. However, point-wise queries of neural fields do not naturally match the point spread function (PSF) of pixels, which may cause aliasing in the super-resolved image. Existing methods attempt to mitigate this by approximating an integral version of the field at each scaling factor, compromising both fidelity and generalization. In this work, we introduce neural heat fields, a novel neural field formulation that inherently models a physically exact PSF. Our formulation enables analytically correct anti-aliasing at any desired output resolution, and -- unlike supersampling -- at no additional cost. Building on this foundation, we propose Thera, an end-to-end ASR method that substantially outperforms existing approaches, while being more parameter-efficient and offering strong theoretical guarantees. The project page is at https://therasr.github.io.