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

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 $360^o$ 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.

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

References in corpus (4)

• Adam: A Method for Stochastic Optimization
• Joint 2D-3D-Semantic Data for Indoor Scene Understanding
• UniFuse: Unidirectional Fusion for 360${}^{\circ }$ Panorama Depth Estimation
• OmniDepth: Dense Depth Estimation for Indoors Spherical Panoramas

Cited by in corpus (8)

• A Survey on RGB-D Datasets
• Hybrid Skip: A Biologically Inspired Skip Connection for the UNet Architecture
• $S^2$Net: Accurate Panorama Depth Estimation on Spherical Surface
• Neural Contourlet Network for Monocular 360 Depth Estimation
• Panoramic Depth Estimation via Supervised and Unsupervised Learning in Indoor Scenes
• Distortion Reduction for Off-Center Perspective Projection of Panoramas
• Sphere-Depth: A Benchmark for Depth Estimation Methods with Varying Spherical Camera Orientations
• A benchmark with decomposed distribution shifts for 360 monocular depth estimation