Geometry Fidelity for Spherical Images

Abstract

Spherical or omni-directional images offer an immersive visual format appealing to a wide range of computer vision applications. However, geometric properties of spherical images pose a major challenge for models and metrics designed for ordinary 2D images. Here, we show that direct application of Fréchet Inception Distance (FID) is insufficient for quantifying geometric fidelity in spherical images. We introduce two quantitative metrics accounting for geometric constraints, namely () and Discontinuity Score (DS). is an extension of FID tailored to additionally capture field-of-view requirements of the spherical format by leveraging cubemap projections. DS is a kernel-based seam alignment score of continuity across borders of 2D representations of spherical images. In experiments, and DS quantify geometry fidelity issues that are undetected by FID.

Cite

Text

Christensen et al. "Geometry Fidelity for Spherical Images." Proceedings of the European Conference on Computer Vision (ECCV), 2024. doi:10.1007/978-3-031-72989-8_16

Markdown

[Christensen et al. "Geometry Fidelity for Spherical Images." Proceedings of the European Conference on Computer Vision (ECCV), 2024.](https://mlanthology.org/eccv/2024/christensen2024eccv-geometry/) doi:10.1007/978-3-031-72989-8_16

BibTeX

@inproceedings{christensen2024eccv-geometry,
  title     = {{Geometry Fidelity for Spherical Images}},
  author    = {Christensen, Anders and Mojab, Nooshin and Patel, Khushman and Ahuja, Karan and Akata, Zeynep and Winther, Ole and Franco, Mar Gonzalez and Colaco, Andrea},
  booktitle = {Proceedings of the European Conference on Computer Vision (ECCV)},
  year      = {2024},
  doi       = {10.1007/978-3-031-72989-8_16},
  url       = {https://mlanthology.org/eccv/2024/christensen2024eccv-geometry/}
}