NBV Background (reading guide)

1 NBV Background

This page is a short literature-oriented entry point. It does not own ARIA-NBV theory, research questions, equations, or implementation contracts. The canonical synthesis is in the Typst thesis, especially its foundations and method sections.

Classical next-best-view work uses geometric uncertainty, information gain, or surface coverage. Learning-based work such as GenNBV [1] and VIN-NBV [2] illustrates the transition from coverage heuristics to quality-oriented view selection. These citations are background evidence only; ARIA-specific claims and the current RRI formulation belong to the thesis.

For the current implementation boundary, see the Python module docstrings and the contract tests under tests/. For the project research questions, use docs/typst/thesis/sections/01-research-questions.typ.

Back to top

References

[1]
X. Chen, Q. Li, T. Wang, T. Xue, and J. Pang, “GenNBV: Generalizable next-best-view policy for active 3D reconstruction,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2024, pp. 16436–16445. Available: https://openaccess.thecvf.com/content/CVPR2024/html/Chen_GenNBV_Generalizable_Next-Best-View_Policy_for_Active_3D_Reconstruction_CVPR_2024_paper.html
[2]
N. Frahm et al., “VIN-NBV: A view introspection network for next-best-view selection.” 2025. Available: https://arxiv.org/abs/2505.06219