Cutting Edge Video Compression

With SVT-AV1-PSY, we believe that we've created an encoder that isn't just capable of competing with the best in the business for AV1, but with newer codecs like VVC as well. The maturity of the x264 encoder allowed it to be a credit to the H.264 codec and improve its perception across the board, and we wish to achieve the same for AV1. The mainline SVT-AV1 project has done incredible work in providing a fast and powerful encoder for AV1, and efforts have been directed toward improving metric scores - our work has focused on producing visually compelling results.

Circuit Board

Visual Comparisons

Our bleeding-edge psychovisual approach has resulted in groundbreaking visual quality that places us far ahead of other open source AV1 encoders. You can use the slider below to compare our performance to other popular encoders, such as x264, x265, aomenc, and VVenC.

Philosophy

Somewhere out there, while we are saying "it isn't rocket science!", there is a rocket scientist saying "it isn't multimedia compression engineering!". Developing video encoders is hard, & requires a deep understanding of human perception. Our approach to video encoding is to leverage our scientific understanding of the human visual system to produce better quality video.

Unlike audio codec development where subjective evaluations are king, video compression is a field ruled by synthetic benchmarks that often fail to capture what is important to our eyes. Metrics are models, and they can be useful, but they are always flawed. While we care about some metrics (like XPSNR, SSIMULACRA2, and Butteraugli), we care more about what you see. Therefore, most of our time goes into understanding how our eyes respond to different compression artifacts and how we can minimize the error our eyes perceive.