用NeRF生成真实感图像，仅靠几十张照片就能训练航天器姿态估计模型。
CAD-Free Learning of Spacecraft Pose Estimators via NeRF-Based Augmentations
- 通过NeRF重建目标，自动生成多样化的视角与光照图像
- 仅需25至400张真实图像即可训练出高精度姿态估计算法
- 适合无完整CAD模型的航天器或复杂光照场景应用
航天器姿态估计模型通常依赖数万张由CAD渲染的图像进行训练。这种对合成数据的依赖（i）限制了在缺乏可靠几何先验的目标（如非合作或文档不全的航天器）上的适用性，（ii）导致在轨真实条件下的泛化性能差，因光照和材质呈现不真实。本文提出一种基于NeRF的图像增强方法，仅需数十至数百张图像即可训练航天器姿态估计算法。该方法学习目标的神经辐射场，通过几何一致的视角与外观增强生成大规模多样化数据集。该增强数据集使无需CAD模型或大规模合成数据集即可训练高精度、特定目标的姿态估计算法成为可能。实验表明，本方法可在仅25至400张真实图像下实现准确姿态估计，即使在极端光照变化条件下亦有效。当应用于大型基于CAD的合成数据集时，该方法还能提升跨域泛化能力，增强对真实在轨条件的鲁棒性。
原文摘要 · Abstract (English)
Spacecraft pose estimation networks require tens of thousands of CAD-rendered images to be trained. This reliance on synthetic CAD data (i) limits applicability to targets with reliable geometry prior, excluding uncooperative or poorly documented spacecraft, and (ii) causes poor generalization to real on-orbit conditions due to unrealistic illumination and material appearance. This paper introduces a NeRF-based image augmentation method that enables the learning of spacecraft pose estimators from only a few tens to a few hundreds of images. The method learns a Neural Radiance Field of the target and generates a large, diverse dataset through geometrically-consistent viewpoint and appearance augmentation. This augmented dataset enables the training of accurate target-specific pose estimators without requiring a CAD model or large synthetic datasets. Experiments show that our approach supports the training of accurate pose estimators from only 25 to 400 realistic images, even under severe illumination variations. When applied on large CAD-based synthetic datasets, the NeRF-based augmentation also enhances out-of-domain generalization, yielding improved robustness to real on-orbit conditions.
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。