SatensoRF: Fast Satellite Tensorial Radiance Field for Multidate Satellite Imagery of Large Size

Existing neural radiance field (NeRF) models for satellite imagery have limitations in processing large images and require solar input, leading to slow speeds. As a response, we introduce SatensoRF, which speeds up the entire process significantly while using fewer parameters for large satellite imagery. We have noticed that the common assumption of Lambertian surfaces in satellite NeRFs (Sat-NeRFs) is not sufficient for vegetative and aquatic elements. In contrast to the traditional hierarchical multilayer perceptron (MLP)-based scene representation, we have chosen a multiscale tensor decomposition approach for color, volume density, and auxiliary variables to model the light field with specular color. Additionally, to rectify inconsistencies in multidate imagery, we incorporate total variation denoising (TVD) to restore the density tensor field, thus mitigating the negative impact of transient objects. To validate our approach, we conducted assessments of SatensoRF using subsets from the SpaceNet multiview dataset, which includes both multidate and single-date multiview RGB images. Our results demonstrate that SatensoRF surpasses the state-of-the-art Sat-NeRF series regarding novel view synthesis performance. Significantly, SatensoRF requires fewer parameters for training, resulting in faster training and inference speeds and reduced computational demands.