Published: 01 Jan 2023, Last Modified: 28 Sept 2024Pattern Recognit. 2023EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract:Highlights•A novel two-stage pose transform network for single image talking head animation.•A novel DIM module with a pose decomposition and cascaded refinement strategy.•A new anime talking head dataset AniHead-2K.