Folks at Adobe have developed a generative AI model that will soon be able to upscale videos by up to eight times their native resolution. In a recently published paper titled VideoGigaGAN, Adobe claims that the model is far superior to traditional video super-resolution (VSR) approaches of the past, which tend to introduce AI artifacts and a whole lot of blurriness in their upscaling attempts.

Thanks to Adobe's generative AI capabilities, "Adobe VideoGigaGAN upsamples videos with high-frequency details while maintaining temporal consistency." Ultimately, the new method produces results with finer-grained details, meaning videos with much higher fidelity and far fewer artifacts.

But don't take our word for it. Here are a few examples that show what Adobe's new VSR method is capable of:

While VSR methods have proven useful in improving images in the past, their success hasn’t translated to video in the same way. And while other upscaling methods attempt to reduce the flicker and other AI weirdness introduced by upsampling, they do so only at the cost of sacrificing detail.