Summary of Jpeg Ai Image Compression Visual Artifacts: Detection Methods and Dataset, by Daria Tsereh et al.
JPEG AI Image Compression Visual Artifacts: Detection Methods and Dataset
by Daria Tsereh, Mark Mirgaleev, Ivan Molodetskikh, Roman Kazantsev, Dmitriy Vatolin
First submitted to arxiv on: 11 Nov 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: Computer Vision and Pattern Recognition (cs.CV); Multimedia (cs.MM)
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 The paper proposes methods to detect, localize, and quantify visual artifacts introduced by neural-network-based image compression methods. The authors focus on three types of artifacts: texture and boundary degradation, color change, and text corruption. They developed a framework to identify regions that are distorted solely due to neural compression but can be recovered successfully by traditional codecs at similar bitrates. The proposed dataset contains 46,440 validated artifacts, collected from the Open Images dataset using different compression-quality parameters. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper is about fixing problems with new ways of compressing images. Sometimes, these new methods make pictures look bad or change them in unwanted ways. The authors found a way to find and measure these problems, so people can use this information to make better image-compression tools. They created a special dataset that has many examples of these problems, which can be used to test and improve neural-network-based image codecs. |
Keywords
» Artificial intelligence » Neural network
