PreSCAN predicts NeRF reconstruction quality in under 30 seconds, achieving a 1000x speedup over Neural Architecture Search.
March 20, 2026
Original Paper
Fast and Generalizable NeRF Architecture Selection for Satellite Scene Reconstruction
arXiv · 2603.18306
The Takeaway
It shifts the bottleneck of satellite 3D reconstruction from hours of GPU-heavy architecture search to near-instant geometric estimation. This allows for real-time deployment on edge devices like Jetson Orin with significantly reduced power and latency.
From the abstract
Neural Radiance Fields (NeRF) have emerged as a powerful approach for photorealistic 3D reconstruction from multi-view images. However, deploying NeRF for satellite imagery remains challenging. Each scene requires individual training, and optimizing architectures via Neural Architecture Search (NAS) demands hours to days of GPU time. While existing approaches focus on architectural improvements, our SHAP analysis reveals that multi-view consistency, rather than model architecture, determines rec