See-through: Single-image Layer Decomposition for Anime Characters
This addresses the tedious manual segmentation and hallucination required in current professional workflows for anime character animation, though it is incremental as it builds on existing methods like diffusion models and Live2D.
The paper tackles the problem of automating the conversion of static anime illustrations into manipulatable 2.5D models by decomposing a single image into inpainted layers with inferred drawing orders, resulting in high-fidelity models suitable for professional animation.
We introduce a framework that automates the transformation of static anime illustrations into manipulatable 2.5D models. Current professional workflows require tedious manual segmentation and the artistic ``hallucination'' of occluded regions to enable motion. Our approach overcomes this by decomposing a single image into fully inpainted, semantically distinct layers with inferred drawing orders. To address the scarcity of training data, we introduce a scalable engine that bootstraps high-quality supervision from commercial Live2D models, capturing pixel-perfect semantics and hidden geometry. Our methodology couples a diffusion-based Body Part Consistency Module, which enforces global geometric coherence, with a pixel-level pseudo-depth inference mechanism. This combination resolves the intricate stratification of anime characters, e.g., interleaving hair strands, allowing for dynamic layer reconstruction. We demonstrate that our approach yields high-fidelity, manipulatable models suitable for professional, real-time animation applications.