{"slug": "fusionrelight-relighting-portraits-in-real-time-via-hybrid-domain-knowledge", "title": "FusionRelight: Relighting Portraits in Real Time via Hybrid Domain Knowledge Fusion", "summary": "Researchers from an undisclosed institution introduced Hybrid Domain Knowledge Fusion (HDKF), a relighting-specific training framework that fuses physics, reflectance, and realism priors from synthetic, One-Light-at-a-Time (OLAT), and in-the-wild data to relight portraits in real time. On a held-out OLAT benchmark, HDKF achieved the best MSE, PSNR, and SSIM among evaluated methods while remaining competitive in LPIPS. The distilled model runs at 512x512 resolution, reaching 11.89 ms on an RTX 2060 and 1.82 ms on an RTX 4090.", "body_md": "Portrait relighting is a low-level vision problem in which physically plausible illumination transfer, identity preservation, and compact real-time inference must be considered together. Iterative diffusion-style methods can synthesize fine detail, but stochastic inference and cost complicate deterministic live video creation; physically grounded relighting preserves identity, but controlled synthetic or light-stage supervision transfers poorly to unconstrained cameras. We present Hybrid Domain Knowledge Fusion (HDKF), a relighting-specific training framework that learns complementary physics, reflectance, and realism priors from synthetic, One-Light-at-A-Time (OLAT), and in-the-wild data, then distills their source-routed supervision into a compact student with clean teacher labels and degraded student inputs. The framework is trained with pixel-aligned RGB, albedo, and normal supervision, providing a simulation substrate for physically grounded low-level relighting. On a held-out OLAT benchmark, HDKF obtains the best MSE, PSNR, and SSIM among evaluated methods while remaining competitive in LPIPS. The distilled model runs in real time at 512x512, reaching 11.89 ms on an RTX 2060 and 1.82 ms on an RTX 4090.", "url": "https://wpnews.pro/news/fusionrelight-relighting-portraits-in-real-time-via-hybrid-domain-knowledge", "canonical_source": "https://research.nvidia.com/publication/2026-08_fusionrelight-relighting-portraits-real-time-hybrid-domain-knowledge-fusion", "published_at": "2026-08-14 16:21:03+00:00", "updated_at": "2026-08-14 16:42:38.826437+00:00", "lang": "en", "topics": ["computer-vision", "generative-ai", "machine-learning"], "entities": ["HDKF", "RTX 2060", "RTX 4090"], "alternates": {"html": "https://wpnews.pro/news/fusionrelight-relighting-portraits-in-real-time-via-hybrid-domain-knowledge", "markdown": "https://wpnews.pro/news/fusionrelight-relighting-portraits-in-real-time-via-hybrid-domain-knowledge.md", "text": "https://wpnews.pro/news/fusionrelight-relighting-portraits-in-real-time-via-hybrid-domain-knowledge.txt", "jsonld": "https://wpnews.pro/news/fusionrelight-relighting-portraits-in-real-time-via-hybrid-domain-knowledge.jsonld"}}