SurGe: Improved Surface Geometry in Point Maps

arXiv (Cornell University)

Abstract

Recent feedforward 3D reconstruction methods predict point maps and estimate global 3D geometry remarkably well. However, their predictions still exhibit inaccurate local surface geometry, which is clearly visible qualitatively but only weakly reflected in common metrics. To make these errors more explicit in evaluation, we introduce a point map normal metric that evaluates the local surface orientation induced by neighboring 3D predictions. To reduce these errors, we propose two complementary components: a point gradient matching loss that supervises depth-normalized 3D finite differences, and a Neighborhood Attention Decoder (NAD) that progressively upsamples features and uses Neighborhood Attention for local feature mixing. Across eight zero-shot monocular geometry benchmarks, our model, SurGe, achieves the best average rank for global point map AbsRel and consistently improves local point map and point map normal evaluations.

Authors 7

1. Karim Knaebel Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University

2. Gonzalo Martín García
3. Christian Schmidt Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University

4. Ilya Fradlin Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University

5. Lucas Nunes Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University

6. Daan de Geus Aachen

   Eindhoven University of Technology · RWTH Aachen University

   Affiliation as printed

   Eindhoven University of Technology https:// vision . rwth-aachen .de/surge

7. Bastian Leibe Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University