
Welcome to MIPG at the Communication University of China!
Our mission is to harness the power of AI for intelligent multimedia processing, covering the expansive terrains of image and video synthesis, treatment, and encoding. Committed to the advancement of Artificial Intelligence Generated Content (AIGC), we are at the forefront of two principal research directions.
Direction 1: AIGC Artistry and Precision in AI-Powered Media Creation
Text-to-Media Intelligence: We transform textual narratives into rich visual stories, transcending traditional storytelling by converting words into compelling images and videos. Conversational AI Content Creation: With the use of expansive language models, we create dynamic dialogues that yield creative content, epitomizing the harmonious collaboration between AI and human ingenuity. Controlled and Personalized Generation: Our technology is tailored not just to fulfill specific conditions but also to embrace user preferences, ensuring each piece of content resonates on a personal level. Intelligent Image Editing: Venturing further, we explore the realm of AI-driven image editing to amplify creative freedom and individual expression.
Direction 2: Intelligent Image and Video Encoding - Crafting the Future of Media Communication
We delve into the intricacies of current image and video encoding techniques, contributing to the research that will define the next-generation standards in intelligent media encoding. Our investigative lens focuses on the symbiotic relationship between generative models and image/video compression, carving out new avenues for generative encoding that promise to redefine media transmission’s efficiency and quality. Join us at MIPG, where innovation meets multimedia, and together we embark on a journey to redefine the digital media landscape through the lens of AI.
We code in the open. We share data. Communicating science requires finding and telling the stories in our data, but these stories are worthless if they don’t stand up to scrutiny from the community.
Based on our media background and technical network resources, we expect to produce high-quality papers and research results in the field of AIGC.
Almost all our projects are done in close collaboration with the worldwide collaborators. Our code and algorithms are designed to solve real scientific problems faced by real users.