Presentation

<
AniDepth : Anime In-between Diffusion using Depth-guided Warped Line-art
SessionML in Production
DescriptionWe propose AniDepth, a novel anime in-betweening method using a video diffusion model enhanced by converting anime illustrations into depth maps. Guided by line-arts, our approach interpolates depth maps and colors to boost fidelity, temporal smoothness, and performance while seamlessly integrating into production pipelines.
Contributors
Sosui Koga
Nara Institute of Science and Technology
Hiroyuki Kubo
Chiba University
Seitaro Shinagawa
Nara Institute of Science and Technology
Yuki Fujimura
Nara Institute of Science and Technology
Kazuya Kitano
Nara Institute of Science and Technology
Akinobu Maejima
OLM Digital, Inc.
IMAGICA GROUP Inc.
Takuya Funatomi
Nara Institute of Science and Technology
Yasuhiro Mukaigawa
Nara Institute of Science and Technology
Event Type
Talk
TimeThursday, 14 August 20252:00pm - 2:22pm PDT
LocationWest Building, Rooms 109-110
Session TimeThursday, 14 August 20252:00pm - 3:30pm PDT
LocationWest Building, Rooms 109-110
Interest Areas
Production & Animation
Recordings
Not Livestreamed
Not Recorded
Keywords
Animation
Generative AI
Next PresentationNext PresentationMachine Learning Meets Lighting: Using Depth Estimation To Build The Light Rigs
Similar Presentations
37. Anime Colorization Using Segment Matching With Candidate Colors
46. Predicting Colors in Unpainted Gaps for Anime-Style Illustration
Towards AI-Driven 3D Creation at the Speed of Thought