SatGeo-NeRF: Geometrically Regularized NeRF for Satellite Imagery

2026-03-23

AI Summary

This work addresses geometric artifacts in satellite-image-based NeRF models caused by overfitting by proposing three model-agnostic geometric regularization techniques: gravity-aligned planarity, granularity control, and depth supervision. These mechanisms enhance cross-ray gradient flow, impose surface normal constraints, and adopt a coarse-to-fine geometric learning strategy, collectively improving geometric consistency and training stability in 3D reconstruction. Evaluated on the DFC2019 benchmark, the proposed approach reduces average elevation error by 13.9% and 11.7% compared to EO-NeRF and EO-GS, respectively, demonstrating a significant improvement in reconstruction accuracy.

Technology Category

• Computer Vision: 3D Computer Vision
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Machine Learning: Deep Neural Architectures and Foundation Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

We present SatGeo-NeRF, a geometrically regularized NeRF for satellite imagery that mitigates overfitting-induced geometric artifacts observed in current state-of-the-art models using three model-agnostic regularizers. Gravity-Aligned Planarity Regularization aligns depth-inferred, approximated surface normals with the gravity axis to promote local planarity, coupling adjacent rays via a corresponding surface approximation to facilitate cross-ray gradient flow. Granularity Regularization enforces a coarse-to-fine geometry-learning scheme, and Depth-Supervised Regularization stabilizes early training for improved geometric accuracy. On the DFC2019 satellite reconstruction benchmark, SatGeo-NeRF improves the Mean Altitude Error by 13.9% and 11.7% relative to state-of-the-art baselines such as EO-NeRF and EO-GS.

Problem

• geometric artifacts
• overfitting
• satellite imagery
• NeRF
• 3D reconstruction

Innovation

• Geometric Regularization
• NeRF
• Satellite Imagery
• Surface Normal Alignment
• Depth Supervision