用快速预测选神经辐射场架构，卫星三维重建提速千倍
Fast and Generalizable NeRF Architecture Selection for Satellite Scene Reconstruction
- 基于几何光度特征预估重建质量，无需训练即可选架构
- 30秒内完成选择，误差小于1分贝，比传统方法快1000倍
- 适合边缘设备部署，降低功耗与延迟，通用性强
神经辐射场（NeRF）在多视角图像的逼真三维重建中表现卓越，但应用于卫星影像仍具挑战：每场景需独立训练，通过神经架构搜索（NAS）优化架构需数小时至数天的GPU时间。我们通过SHAP分析发现，重建质量主要由多视角一致性决定，而非模型架构。基于此，提出PreSCAN框架，利用轻量级几何与光度描述符，在训练前快速预测NeRF质量。PreSCAN可在30秒内完成架构选择，预测误差低于1分贝，相较NAS实现1000倍加速。进一步在Jetson Orin边缘平台验证，结合离线成本分析可使推理功耗降低26%，延迟减少43%，质量损失极小。DFC2019数据集实验表明，PreSCAN无需重训即可跨多样化卫星场景泛化。
原文摘要 · Abstract (English)
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 reconstruction quality. Based on this insight, we develop PreSCAN, a predictive framework that estimates NeRF quality prior to training using lightweight geometric and photometric descriptors. PreSCAN selects suitable architectures in < 30 seconds with < 1 dB prediction error, achieving 1000$\times$ speedup over NAS. We further demonstrate PreSCAN's deployment utility on edge platforms (Jetson Orin), where combining its predictions with offline cost profiling reduces inference power by 26% and latency by 43% with minimal quality loss. Experiments on DFC2019 datasets confirm that PreSCAN generalizes across diverse satellite scenes without retraining.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。