Recently, the review results for the ACM SIGGRAPH 2025 (International Conference on Computer Graphics and Interactive Techniques) were announced. Art papers from the team led by Associate Professor Xiang Fan of the Department of Visual Communication Design at the Academy of Arts & Design, Tsinghua University were accepted by the conference. This marks the third time the research team has been recognized, following their Best Art Paper Award in 2020. This achievement fully demonstrates their sustained innovation and consistent academic contributions in this research field.
The SIGGRAPH annual conference is not only a highly authoritative and influential international conference in the field of computer graphics, but it also features research outcomes across various domains including artificial intelligence, machine learning, image processing, and video. It consistently offers new research directions and inspires innovation across both academia and industry.
An AI art paper, supervised by Associate Professor Xiang Fan and created by Yu Tongge, a graduate student from the Tsinghua University-University of Washington Global Innovation Exchange (GIX) program, is slated for presentation at the SIGGRAPH 2025 Art Papers program in Vancouver, Canada, in August 2025. It will also be published in a special issue of the Proceedings of the ACM on Computer Graphics and Interactive Techniques (PACM CGIT). This project leverages historical audio cassette archives of Shenzhen radio’s late-night programs from the 1990s. Combining these with artificial intelligence technology, the team developed an interactive chat program that allows users to traverse a 30-year time tunnel and engage in live calls with a digital clone of the original radio host. Users simply dial a hotline number on their mobile devices to interact in real-time with the program’s digital host, sharing youthful concerns and experiencing the popular music, fashion, and trends of the 90s.
This project endeavors to transcend time and space to awaken fading collective urban memories, directly addressing the current international research hotspot of integrating humanities and technology. It received high praise from international reviewers: “Through AI voice cloning and Large Language Models (LLMs), re-emerging in the form of a digital twin of the host’s voice, the project preserves authentic sound effects and invaluable memories. The paper clearly articulates the broader broadcasting cultural context, highlighting the significance of radio programs in the 1990s as a force in shaping cultural identity.”
In 2020, Associate Professor Xiang Fan led her team to win the SIGGRAPH Best Art Paper Award for their work, Augmented Family Tree. This project utilized the China Biographical Database (CBDB), primarily developed by Harvard University, to explore the millennia-long origins of Chinese lineage, offering clues to the question “Where do I come from?” through visualization technology. Following this, she was invited to serve as an Art Papers Review Committee member for SIGGRAPH 2021 and SIGGRAPH Asia 2023.
In 2022, Kevin Maher, an American Master’s student from the Department of Visual Communication Design whom she supervised, had his project FaceType selected for the SIGGRAPH Art Long Papers program. Inspired by Chinese calligraphy and developed in collaboration with the Chinese Academy of Sciences, this project generated “emotional calligraphy” from facial expressions, integrating the audience’s nuanced emotions into the forms of English alphabet letters. The project also received the Tsinghua University Outstanding Master’s Thesis Award that same year.
“FaceType” Project Introduction
Moving forward, the research team will continue to delve into areas such as cultural memory, artificial intelligence, and interactive art, tracing the path back through the tunnel of time to seek out experiences worth revisiting.
