Octree-based hierarchical sampling optimization for the volumetric super-resolution of scientific data

Xinjie Wang, Maoquan Sun, Yundong Guo, Chunxin Yuan, Xiang Sun, Zhiqiang Wei, Xiaogang Jin

Abstract: Highlights•We propose a novel octree-based approach for modeling volumetric scientific data by leveraging its importance distribution.•We present a hierarchical sampling method to optimize the distribution of sample points in each training data block.•We construct a multi-stage training strategy to avoid failure of the proposed sampling optimization.