Lecture: On the Training-Free Long Video Generation

Abstract: I will present FIFO-Diffusion, a novel approach for generating infinitely long videos using a pretrained short video generation model. This method enables seamless video synthesis by jointly denoising frames with varying noise levels, effectively bridging the gap between training and inference. Its design is particularly well-suited for long video generation, addressing temporal consistency challenges that arise during extrapolation. Additionally, I will briefly discuss recent advancements in training-free image-to-image translation, focusing on text-driven image editing strategies, by leveraging pretrained diffusion models. These approaches offer new perspectives on efficient generative modeling, expanding the accessibility and scalability of diffusion-based synthesis.

Bio: Bohyung Han is a Professor in the Department of Electrical and Computer Engineering at Seoul National University, Korea. He received his Ph.D. from the Department of Computer Science at the University of Maryland, College Park, MD, USA, in 2005. He has served or will serve as an organizing and senior-level program committee member in major computer vision and machine learning conferences numerous times including a Program Chair in ICCV 2025 and Senior Area Chair in NeurIPS, CVPR, ICLR, and ICML. He is also an Associate Editor of TPAMI. He received the Google AI Focused Research Award in 2018. His research interests include computer vision and machine learning with an emphasis on deep learning.