This 3-hours tutorial is for beginners who would like to quickly get into Video Diffusion Models (VDMs), covering study various subtopics of VDMs and their corresponding works as of Nov 2023. Due to the limited time for preparing this tutorial over the past several weeks, likely we miss out some works, sorry for that. Please feel free to command in the below and add some introduction about the paper, thank you so much.
Slides pdf download: sites.google.com/view/showlab/tutorial
Speaker Bio: www.comp.nus.edu.sg/cs/people/mikeshou/
Twitter: MikeShou1
Outline:
00:00:00 - 0. Intro & outline
00:03:09 - 1. Fundamentals of Diffusion Models
00:22:28 - 2.1 Video Gen, Pioneering Works, Task Definition and Evaluation Metrics
00:51:32 - 2.2 Video Gen, Open-source Base Models
01:09:50 - 2.3 Video Gen, Other Closed-source Works
01:10:12 - 2.4 Video Gen, Training-Efficient Techniques
01:19:56 - 2.5 Video Gen, Storyboard
01:35:51 - 2.6 Video Gen, Long Video Generation
01:40:08 - 2.7 Video Gen, Multimodal-Guided
01:45:23 - 3.1 Video Edit, Tuning-Based
02:08:23 - 3.2 Video Edit, Training-Free
02:14:40 - 3.3 Video Edit, Controlled Editing
02:56:36 - 3.4 Video Edit, 3D-Aware
03:15:52 - 3.5 Video Edit, Other Guidance
03:19:50 - 4. Summary
#aigc #videogeneration #videoediting #ai #computervision