View-Dependent Octree-based Mesh Extraction in Unbounded Scenes for Procedural Synthetic Data
This addresses the need for high-quality procedural synthetic data in computer vision, though it appears incremental as it builds on existing SDF-based methods.
The paper tackles the problem of mesh extraction from procedural signed distance functions (SDFs) in unbounded scenes, which often has artifacts or performance issues, and proposes OcMesher to achieve efficient, high-detail extraction with perfect view-consistency for synthetic data generation.
Procedural synthetic data generation has received increasing attention in computer vision. Procedural signed distance functions (SDFs) are a powerful tool for modeling large-scale detailed scenes, but existing mesh extraction methods have artifacts or performance profiles that limit their use for synthetic data. We propose OcMesher, a mesh extraction algorithm that efficiently handles high-detail unbounded scenes with perfect view-consistency, with easy export to downstream real-time engines. The main novelty of our solution is an algorithm to construct an octree based on a given SDF and multiple camera views. We performed extensive experiments, and show our solution produces better synthetic data for training and evaluation of computer vision models.