English

Pano3D: A Holistic Benchmark and a Solid Baseline for $360^o$ Depth Estimation

Computer Vision and Pattern Recognition 2021-09-08 v1 Machine Learning

Abstract

Pano3D is a new benchmark for depth estimation from spherical panoramas. It aims to assess performance across all depth estimation traits, the primary direct depth estimation performance targeting precision and accuracy, and also the secondary traits, boundary preservation, and smoothness. Moreover, Pano3D moves beyond typical intra-dataset evaluation to inter-dataset performance assessment. By disentangling the capacity to generalize to unseen data into different test splits, Pano3D represents a holistic benchmark for 360o360^o depth estimation. We use it as a basis for an extended analysis seeking to offer insights into classical choices for depth estimation. This results in a solid baseline for panoramic depth that follow-up works can build upon to steer future progress.

Keywords

Cite

@article{arxiv.2109.02749,
  title  = {Pano3D: A Holistic Benchmark and a Solid Baseline for $360^o$ Depth Estimation},
  author = {Georgios Albanis and Nikolaos Zioulis and Petros Drakoulis and Vasileios Gkitsas and Vladimiros Sterzentsenko and Federico Alvarez and Dimitrios Zarpalas and Petros Daras},
  journal= {arXiv preprint arXiv:2109.02749},
  year   = {2021}
}

Comments

Presented at the OmniCV CVPR 2021 workshop. Code, models, data and demo at https://vcl3d.github.io/Pano3D/

R2 v1 2026-06-24T05:44:10.882Z