{"slug": "geometry-driven-shadow-harmonisation-for-composited-faces-a-multiplicative", "title": "Geometry-Driven Shadow Harmonisation for Composited Faces: A Multiplicative, Albedo-Preserving Relighting Pipeline", "summary": "A new arXiv paper (arXiv:2609.17740v1) presents a geometry-driven form-shadow injection pipeline for composited faces that multiplies a per-pixel gain field onto linear RGB rather than repainting the face, preserving albedo, identity, and skin tone. On an analytic face heightfield, the default three-parameter transfer (τ, σ, g_min) = (0.90, 0.45, 0.82) modifies 56.5% of face pixels with a mean gain of 0.938 (0.890 on modified pixels) and drives 3.4% of pixels to the floor. The authors analyse the transfer in closed form, ablate its parameters, and discuss failure modes of the monotone, darkening-only formulation, including double-shadowing of non-flat donors.", "body_md": "arXiv:2609.17740v1 Announce Type: new \nAbstract: Face swapping and face compositing pipelines routinely produce a face that is geometrically well aligned but photometrically implausible: the donor face carries flat, near-frontal studio illumination while the host body and background carry directional scene light. Most existing remedies re-synthesise the face through colour transfer, neural relighting, or inverse rendering, and therefore risk altering identity, skin tone, and texture. We present a conservative alternative: geometry-driven form-shadow injection. The pipeline never repaints the face. It estimates a per-pixel gain field $g\\in[g_{\\min},1]$ from a rasterised 3D face proxy and multiplies it channel-uniformly onto linear RGB, so the operator can only darken and cannot shift chromaticity. A dense landmark mesh is rasterised into a depth buffer, from which we derive surface normals, a cavity term, and screen-space cast shadows. Key-light direction is estimated from host-side cues (body, background, hair halo); on-face cues are downweighted because they recover the donor's lighting. Shadow magnitude is not matched to the host: it is set by a three-parameter transfer $(\\tau,\\sigma,g_{\\min})$. The shading field is divided by its 75th percentile over skin, then gated, scaled, clamped, smoothed, and re-clipped inside a feathered, skin-gated face mask. On an analytic face heightfield, the default $(\\tau,\\sigma,g_{\\min})=(0.90,0.45,0.82)$ modifies 56.5% of face pixels with mean gain 0.938 (0.890 on modified pixels) and drives 3.4% of pixels to the floor. Hue invariance is a corollary of the operator. We analyse the transfer in closed form, ablate its parameters, and discuss failure modes of a monotone, darkening-only formulation, including double-shadowing of non-flat donors.", "url": "https://wpnews.pro/news/geometry-driven-shadow-harmonisation-for-composited-faces-a-multiplicative", "canonical_source": "https://arxiv.org/abs/2609.17740", "published_at": "2026-09-17 04:00:00+00:00", "updated_at": "2026-09-17 04:26:37.774697+00:00", "lang": "en", "topics": ["computer-vision", "generative-ai", "ai-research"], "entities": ["arXiv", "arXiv:2609.17740v1"], "alternates": {"html": "https://wpnews.pro/news/geometry-driven-shadow-harmonisation-for-composited-faces-a-multiplicative", "markdown": "https://wpnews.pro/news/geometry-driven-shadow-harmonisation-for-composited-faces-a-multiplicative.md", "text": "https://wpnews.pro/news/geometry-driven-shadow-harmonisation-for-composited-faces-a-multiplicative.txt", "jsonld": "https://wpnews.pro/news/geometry-driven-shadow-harmonisation-for-composited-faces-a-multiplicative.jsonld"}}