Get the latest tech news
Arbitrary-Scale Super-Resolution with Neural Heat Fields
Thera is a super-resolution approach that leverages neural heat fields for anti-aliased results.
The components are then linearly combined using coefficients $\mathbf{W}_2$, resulting in an appropriately-blurred, continuous local neural field. This field is then rasterized at the appropriate sampling rate (resolution) to yield a part of the final output image (red square). Due to its principled observation model, our method achieves state-of-the-art performance on a variety of super-resolution benchmarks.
Or read this on Hacker News
