The author

PhotoAIBench is written and maintained by one person, a Principal ML engineer, who publishes under the site's name.

Why this site

Computer vision is my core field. I was the lead ML platform engineer for a computer-vision platform that ran on 5,000 live video streams, and I created Videoflow, an open-source framework for building video-analysis pipelines. Measuring what a vision model does to pixels, and noticing when a metric and the picture disagree, is everyday work in that job. PhotoAIBench applies it to the photo tools people actually use: upscalers, background removers and restoration apps.

Most "best AI upscaler" articles show one flattering example per tool and no method. PhotoAIBench works the other way round: the test photos are public and free to use, the method is published before the results, the scores come from code you can run, and the free PSNR and SSIM calculator uses the same formulas so you can check any number on your own images.

How the site is funded

The tools and the test results are free. The site may earn affiliate commissions from some photo-software companies. Every such link will be labeled "(paid link)" where it appears; today none are active, so every vendor link is a plain link. The affiliate disclosure lists every relationship and the rules that keep commissions out of the results. There is no advertising and no sponsored content.

Independence

PhotoAIBench is not affiliated with any company whose tools it tests. Companies cannot pay to be tested, ranked, described or left out, and the site publishes no reviews or testimonials. A tool is tested only within its own terms of service, and a company sees its results before they are published (see the methodology).

Contact

Email [email protected] to report a wrong license, source or number, or a tool whose terms or free tier have changed. Corrections are dated on the page they affect.

Also available as Markdown.