{"slug": "domain-tuned-diffusion-super-resolution-for-4k-facial-asset-maps", "title": "Domain-Tuned Diffusion Super-Resolution for 4K Facial Asset Maps", "summary": "A domain-tuned diffusion upscaler converts 1K facial asset maps to 4K (1024 to 4096 pixels) by adapting a frozen SDXL backbone with lightweight LoRA modules per modality while a Tile ControlNet preserves input structure, according to an arXiv paper (2611.21890v1). On held-out maps the method reduced albedo LPIPS from 0.21 to 0.08 and produced sharper, more faithful detail than bilinear and off-the-shelf GAN and diffusion super-resolvers, while avoiding misplaced or excessive displacement relief. The authors conclude that lightweight domain adaptation can yield render-ready 4K facial asset maps without a custom super-resolution architecture.", "body_md": "arXiv:2611.21890v1 Announce Type: new \nAbstract: Physically based facial assets require 4K albedo and displacement maps for cinematic close-ups, yet generative pipelines commonly produce them at only 1K. Naively applying photographic super-resolution is a poor fit: facial UV maps have domain-specific statistics, and displacement is a scalar height field whose errors directly distort geometry. We present a domain-tuned diffusion upscaler for 4x (1024 to 4096) facial asset super-resolution. Starting from a bilinearly upsampled map, we adapt a frozen SDXL backbone per modality with lightweight LoRA modules, while a Tile ControlNet preserves input structure as the diffusion prior restores pore- and wrinkle-scale detail. On held-out maps, our method produces sharper, more faithful detail than bilinear and off-the-shelf GAN and diffusion super-resolvers, reducing albedo LPIPS from 0.21 to 0.08 while avoiding misplaced or excessive displacement relief. These results show that lightweight domain adaptation can produce render-ready 4K facial asset maps without a custom super-resolution architecture.", "url": "https://wpnews.pro/news/domain-tuned-diffusion-super-resolution-for-4k-facial-asset-maps", "canonical_source": "https://arxiv.org/abs/2611.21890", "published_at": "2026-11-18 05:00:00+00:00", "updated_at": "2026-11-18 05:15:11.793555+00:00", "lang": "en", "topics": ["generative-ai", "computer-vision", "ai-research", "machine-learning"], "entities": ["SDXL", "LoRA", "Tile ControlNet", "arXiv"], "also_reported_by": [], "takeaway": "Adapting a frozen SDXL backbone with per-modality LoRA modules plus a Tile ControlNet upscales facial albedo and displacement maps 4x from 1024 to 4096, cutting albedo LPIPS from 0.21 to 0.08 versus bilinear and off-the-shelf super-resolvers.", "recommendation": null, "alternates": {"html": "https://wpnews.pro/news/domain-tuned-diffusion-super-resolution-for-4k-facial-asset-maps", "markdown": "https://wpnews.pro/news/domain-tuned-diffusion-super-resolution-for-4k-facial-asset-maps.md", "text": "https://wpnews.pro/news/domain-tuned-diffusion-super-resolution-for-4k-facial-asset-maps.txt", "jsonld": "https://wpnews.pro/news/domain-tuned-diffusion-super-resolution-for-4k-facial-asset-maps.jsonld"}}