Presentation / Installation


35. SegAnimeChara: Segmenting Anime Characters Generated by AI
SessionPosters: Image, Video, & Computer Vision
DescriptionThis work introduces SegAnimeChara, a novel system of transforming AI-generated anime images into game characters while retaining unique features. SegAnimeChara can efficiently, zero-shot, segment anime body parts from generative images. Furthermore, the system conserves the game character's unique outfit and reduces the redundant duplicate text prompts in semantic segmentation.

Want to hear more? Join the interactive discussion with the poster contributor at the end of this Technical Papers session:

Image and Video Editing
Monday, 7 August, 2-3:30 pm, Petree Hall D
Contributors
Andy Yu-Hsiang Tseng
National Taiwan University
Wen-Fan Wang
National Taiwan University
Bing-Yu Chen
National Taiwan University
Event Type
Poster
TimeThursday, 10 August 20239am - 5:30pm PDT
LocationWest Lobby Entrance
ACM Digital Library Poster PDF
Session Time & Location
Sunday, 6 August 20239am - 5:30pm PDTWest Lobby Entrance
Monday, 7 August 20239am - 5:30pm PDTWest Lobby Entrance
Tuesday, 8 August 20239am - 5:30pm PDTWest Lobby Entrance
Wednesday, 9 August 20239am - 5:30pm PDTWest Lobby Entrance
Thursday, 10 August 20239am - 5:30pm PDTWest Lobby Entrance
Keywords
Adaptability
Registration Categories
FC
FCS
E
EFC
EE
Next PresentationNext Presentation
36. Smart Scaling: A Hybrid Deep-Learning Approach to Content-Aware Image Retargeting