Summary of Boosting Flow-based Generative Super-resolution Models Via Learned Prior, by Li-yuan Tsao et al.
Boosting Flow-based Generative Super-Resolution Models via Learned Prior
by Li-Yuan Tsao, Yi-Chen Lo, Chia-Che Chang, Hao-Wei Chen, Roy Tseng, Chien Feng, Chun-Yi Lee
First submitted to arxiv on: 16 Mar 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This work introduces a conditional learned prior to the inference phase of flow-based super-resolution (SR) models, addressing challenges such as grid artifacts and suboptimal results. The proposed framework integrates seamlessly with any contemporary flow-based SR model without modifying its architecture or pre-trained weights. It predicts a latent code conditioned on the low-resolution image, which is then transformed by the flow model into an SR image. Extensive experiments and ablation analyses demonstrate the effectiveness of this framework in enhancing the performance of flow-based SR models. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper solves problems with super-resolution (SR) images. Right now, making high-quality SR images is tricky because it can look blocky or not perfect. The researchers created a new way to make SR images that works better by adding information from the low-resolution image before it’s transformed into a high-resolution image. This helps fix issues like blocky areas and makes the final result better. |
Keywords
» Artificial intelligence » Inference » Super resolution
