Sphere-Depth: A Benchmark for Depth Estimation Methods with Varying Spherical Camera Orientations

2026-04-25

AI Summary

Existing monocular depth estimation methods suffer significant performance degradation on spherical images due to camera pose variations and distortions inherent in equirectangular projection, compounded by the absence of a systematic evaluation benchmark. This work proposes Sphere-Depth, the first comprehensive benchmark and calibration protocol specifically designed for depth estimation on spherical images under pose perturbations. By simulating non-standard camera poses, introducing a unified depth-calibrated error metric, and conducting extensive evaluations across diverse models—including Depth Anything, BiFuse++, and SliceNet—the study reveals that even spherical-aware architectures exhibit severe performance degradation under pose shifts. The project code and benchmark are publicly released to facilitate future research.

Technology Category

• Computer Vision: Biometrics, Face, Gesture & Pose
• Machine Learning: Calibration & Uncertainty Quantification
• Intelligent Robots: State Estimation

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Reliable depth estimation from spherical images is crucial for 360° vision in robotic navigation and immersive scene understanding. However, the onboard spherical camera can experience unintentional pose variations in real-world robotic platforms that, along with the geometric distortions inherent in equirectangular projections, significantly impact the effectiveness of depth estimation. To study this issue, a novel public benchmark, called Sphere-Depth, is introduced to systematically evaluate the robustness of monocular depth estimation models from equirectangular images in a reproducible way. Camera pose perturbations are simulated and used to assess the performance of a popular perspective-based model, Depth Anything, and of spherical-aware models such as Depth Anywhere, ACDNet, Bifuse++, and SliceNet. Furthermore, to ensure meaningful evaluation across models, a depth calibration-based error protocol is proposed to convert predicted relative depth values into metric depth values using supervised learned scaling factors for each model. Experiments show that even models explicitly designed to process spherical images exhibit substantial performance degradation when variations in the camera pose are observed with respect to the canonical pose. The full benchmark, evaluation protocol, and dataset splits are made publicly available at: https://github.com/sgazzeh/Sphere_depth

Problem

• depth estimation
• spherical images
• camera orientation
• pose variation
• equirectangular projection

Innovation

• spherical depth estimation
• camera pose perturbation
• equirectangular projection
• depth calibration
• robustness benchmark