PULSE is a self-supervised photo upsampling algorithm. Instead of starting with the LR image and slowly adding detail, PULSE traverses the high-resolution natural image manifold, searching for images that downscale to the original LR image. This is formalized through the downscaling loss, which guides exploration through the latent space of a generative model. By leveraging properties of high-dimensional Gaussians, the authors aim to restrict the search space to guarantee realistic outputs.
Source: PULSE: Self-Supervised Photo Upsampling via Latent Space Exploration of Generative ModelsPaper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Super-Resolution | 2 | 28.57% |
Heart rate estimation | 1 | 14.29% |
Photoplethysmography (PPG) | 1 | 14.29% |
Denoising | 1 | 14.29% |
Face Hallucination | 1 | 14.29% |
Image Super-Resolution | 1 | 14.29% |
Component | Type |
|
---|---|---|
🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |