Domain-Tuned Diffusion Super-Resolution for 4K Facial Asset Maps
Organizations: Electronic Arts Vancouver, Canada · Electronic Arts Montréal, Canada · Electronic Arts Toronto, Canada
Abstract
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.
Figures & tables
| Albedo | Displacement | ||||||
|---|---|---|---|---|---|---|---|
| Method | PSNR | SSIM | LPIPS | PSNR | SSIM | Detail ( ) | Corr |
| Bilinear | 35.69 | 0.94 | 0.21 | 36.22 | 0.93 | 0.46 | 0.61 |
| Real-ESRGAN ( Wang et al., 2021 ) | 29.69 | 0.86 | 0.25 | 31.12 | 0.81 | 1.26 | 0.33 |
| SD ( ) ( Rombach et al., 2022 ) | 31.99 | 0.77 | 0.49 | 32.57 | 0.81 | 1.15 | 0.36 |
| Ours (fine-tuned) | 35.72 | 0.94 | 0.08 | 35.91 | 0.91 | 0.97 | 0.61 |
Image Super-Resolution
Image super-resolution reconstructs high-resolution images from lower-resolution observations, balancing detail, fidelity, and computational cost.
See all 274 papers