{"slug": "bunraku-turning-a-single-illustration-into-an-editable-live2d-character", "title": "Bunraku: Turning a Single Illustration into an Editable Live2D Character", "summary": "Researchers introduced Bunraku, the first system that generates a complete Live2D character model from a single illustration, including ordered RGBA layers, deformation meshes, and keypose vertex offsets. On 50 held-out characters, Stage 2 achieved a per-vertex direction cosine of 0.768 (median 0.828). The team also released Live2D-Bench, the first standardized benchmark, and an 8,884-model corpus with layer and animation supervision.", "body_md": "arXiv:2607.27348v1 Announce Type: new\nAbstract: Live2D is the dominant 2D character-animation format for anime characters and virtual avatars, representing each character as a stack of RGBA layers driven by per-layer mesh deformation. Despite its wide use in virtual streaming, mobile games, and interactive characters, authoring a Live2D model still demands weeks of manual layer separation, occlusion completion, mesh placement, and keyframing, and no prior generative method produces such a structured asset end-to-end. We present the first system that, from a single illustration, generates all the structured information a Live2D runtime consumes: ordered RGBA layers, a deformation mesh per layer, and the parameter-driven keypose vertex offsets that make the character move. Stage 1 casts layered decomposition as a layered diffusion process under a Live2D-aware organ-level taxonomy, producing an ordered RGBA stack with hidden-region completion. Stage 2 builds a content-conforming triangle mesh for each layer from its alpha channel alone, then predicts the keypose displacement field of all layers jointly: every vertex of every layer is one token, self-attention spans layer boundaries, and each displacement is factorised into a bounded direction and a log-magnitude. Joint rather than independent prediction is what makes the result a coherent character instead of separately plausible parts, and is our largest gain; scaling the network 112x yields none. On 50 held-out characters, under true generation with no teacher forcing, Stage 2 attains a per-vertex direction cosine of 0.768 (median 0.828). Because a layer's mesh derives from its alpha channel, a clothing layer can be re-textured from a natural-language instruction while the mesh and predicted animation are reused byte-for-byte. We further contribute Live2D-Bench, the first standardized benchmark for the task, and an 8,884-model Live2D corpus with layer and animation supervision.", "url": "https://wpnews.pro/news/bunraku-turning-a-single-illustration-into-an-editable-live2d-character", "canonical_source": "https://arxiv.org/abs/2607.27348", "published_at": "2026-07-31 04:00:00+00:00", "updated_at": "2026-07-31 04:38:55.669465+00:00", "lang": "en", "topics": ["artificial-intelligence", "generative-ai", "computer-vision"], "entities": ["Bunraku", "Live2D", "Live2D-Bench"], "alternates": {"html": "https://wpnews.pro/news/bunraku-turning-a-single-illustration-into-an-editable-live2d-character", "markdown": "https://wpnews.pro/news/bunraku-turning-a-single-illustration-into-an-editable-live2d-character.md", "text": "https://wpnews.pro/news/bunraku-turning-a-single-illustration-into-an-editable-live2d-character.txt", "jsonld": "https://wpnews.pro/news/bunraku-turning-a-single-illustration-into-an-editable-live2d-character.jsonld"}}