Collaborating with experts from Google and Google DeepMind, 6G@UT Austin faculty members Sanjay Shakkottai, Constantine Caramanis, and PhD student Litu Rout have developed cutting-edge algorithms that improve image editing, restoration, and personalization processes. This approach not only paves the way for enhanced image manipulation capabilities but also showcases the versatility of latent representation spaces in solving complex visual challenges such as inpainting, denoising, deblurring, de-striping, and super resolution. Production and development teams at Google are actively integrating these algorithms into next-generation devices and production pipelines, including Pixel, Chromebook, Tablet, and YouTube. As generative AI continues to evolve, these findings are poised to significantly impact the fields of computer vision and digital content creation.
A recent article published by IFML explores the research in depth - read the full article here. In addition, Dr. Shakkottai published a tutorial as part of the IFML Diffusion Seminar Series - see the tutorial here.