用分块+重叠策略，让单张显卡跑通全球卫星3D重建。
Tile and Slide : A New Framework for Scaling NeRF from Local to Global 3D Earth Observation
- 分块处理不重叠的3D区域，每块独立训练避免内存爆炸。
- 在单张GPU上实现线性时间复杂度，重建质量无损失。
- 新设计的2×2分块推进与采样器，消除块边拼接误差。
神经辐射场（NeRF）最近成为从多视角卫星图像进行3D重建的新范式。然而，当前最先进的NeRF方法通常受限于小场景，因其训练时内存占用过高。本文研究了这一问题，提出Snake-NeRF框架以实现大场景扩展。我们采用外存方法，无需同时加载所有图像和网络，可在单一设备上运行。通过将兴趣区域划分为不重叠的3D瓦片，并对图像进行重叠裁剪，确保每个瓦片训练时包含全部必要像素。引入新颖的$2\times 2$ 3D瓦片推进策略与分段采样器，共同防止瓦片边界处的3D重建误差。实验表明，大型卫星图像可在线性时间复杂度下，仅用单张GPU完成处理，且质量无损。
原文摘要 · Abstract (English)
Neural Radiance Fields (NeRF) have recently emerged as a paradigm for 3D reconstruction from multiview satellite imagery. However, state-of-the-art NeRF methods are typically constrained to small scenes due to the memory footprint during training, which we study in this paper. Previous work on large-scale NeRFs palliate this by dividing the scene into NeRFs. This paper introduces Snake-NeRF, a framework that scales to large scenes. Our out-of-core method eliminates the need to load all images and networks simultaneously, and operates on a single device. We achieve this by dividing the region of interest into NeRFs that 3D tile without overlap. Importantly, we crop the images with overlap to ensure each NeRFs is trained with all the necessary pixels. We introduce a novel $2\times 2$ 3D tile progression strategy and segmented sampler, which together prevent 3D reconstruction errors along the tile edges. Our experiments conclude that large satellite images can effectively be processed with linear time complexity, on a single GPU, and without compromise in quality.
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。