EUVIMEDEUVIMED entdecken
Crossref

UltraNormal: Scaling Pretrained Diffusion‐Based Surface Normal Estimators to Ultra‐High‐Resolution

Satyam Bhardwaj, Lokender Tiwari, Shivam Ashok Shukla, Mritunjoy Halder, Aalok Gangopadhyay, Brojeshwar Bhowmick

Computer Graphics Forum · 05.08.2026

DOI: 10.1111/cgf.70587 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Computer Graphics Forum, online erschienen am 05.08.2026. Verfasst von Satyam Bhardwaj und 5 weiteren Autor:innen. Die Arbeit nennt 47 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Most of the current diffusion‐based surface normal estimators were trained at low resolutions (e.g., ), significantly below the resolution of modern cameras. Applying such models to high‐resolution images causes two issues: (1) downsampling removes high‐frequency geometric details, and (2) direct high‐resolution inference loses global context. To address these challenges, we propose UltraNormal, which augments existing pretrained diffusion‐based surface normal estimators by using overlapping sliding patches with test‐time guidance to inject global context. UltraNormal initializes each patch with the latent of the corresponding low‐resolution estimate enhanced with high‐frequency details from the local estimate and resumes the reverse diffusion process from an intermediate time step. This approach maintains geometric accuracy of the global prediction while adding high‐frequency details without retraining the diffusion model. To the best of our knowledge, this is the first test‐time method to scale low‐resolution diffusion‐based surface normal estimators to 4K resolution without any retraining. We build UltraNormal upon three different base models and evaluate on both real‐world data and a 4K photorealistic synthetic dataset. We perform extensive ablations and demonstrate improved accuracy, sharpness, and surface reconstruction quality compared to the base model.

Zitieren

APA

Bhardwaj, S., Tiwari, L., Shukla, S. A., Halder, M., Gangopadhyay, A., & Bhowmick, B. (2026). UltraNormal: Scaling Pretrained Diffusion‐Based Surface Normal Estimators to Ultra‐High‐Resolution. Computer Graphics Forum. https://doi.org/10.1111/cgf.70587

Vancouver

Bhardwaj S, Tiwari L, Shukla SA, Halder M, Gangopadhyay A, Bhowmick B. UltraNormal: Scaling Pretrained Diffusion‐Based Surface Normal Estimators to Ultra‐High‐Resolution. Computer Graphics Forum. 2026. doi:10.1111/cgf.70587

BibTeX
@article{Bhardwaj2026,
  author = {Bhardwaj, Satyam and Tiwari, Lokender and Shukla, Shivam Ashok and Halder, Mritunjoy and Gangopadhyay, Aalok and Bhowmick, Brojeshwar},
  title = {UltraNormal: Scaling Pretrained Diffusion‐Based Surface Normal Estimators to Ultra‐High‐Resolution},
  journal = {Computer Graphics Forum},
  year = {2026},
  doi = {10.1111/cgf.70587}
}
Kostenlos starten →