T3lescope: Arbitrary-Resolution High-Fidelity Generative Surface Reconstruction from Images

Key points

• Per-scene optimization methods lack the learned 3D prior needed when observations are sparse or surfaces are glossy or transparent.1
• To address these issues, we propose T3lescope, which applies a single fixed-resolution generator across scene scales in an inference-time coarse-to-fine cascade.1
• A coarse level establishes the scene layout, and finer levels perturb and denoise geometry inherited from the coarser level within progressively finer spatial cells to recover surface detail.1
• On indoor, outdoor, and city-scale scenes, T3lescope outperforms feed-forward and generative baselines, matches or surpasses per-scene optimization, and recovers fine structures as well as glossy and transparent surfaces.1

Sources (1)

• [1]T3lescope: Arbitrary-Resolution High-Fidelity Generative Surface Reconstruction from Images
  Hugging Face Daily Papers · Oct 2, 12:00 AM

  We reconstruct high-fidelity 3D scene meshes from posed multi-view images without per-scene optimization, across scales ranging from single objects to large outdoor scenes.

  Per-scene optimization methods lack the learned 3D prior needed when observations are sparse or surfaces are glossy or transparent.

Extractive summary: sentences quoted from the sources.