Inferring bias and uncertainty in camera calibration

TL;DR

This paper presents a comprehensive evaluation scheme for camera calibration that detects biases and estimates uncertainty, improving calibration accuracy assessment and enabling benchmarking of calibration methods.

Contribution

It introduces new bias detection and uncertainty estimation methods that extend classical approaches and are applicable across different camera models and calibration setups.

Findings

Bias detection reveals calibration setup imperfections.

A resampling-based uncertainty estimator works under non-ideal conditions.

The methods are validated with simulations and real camera data.

Abstract

Accurate camera calibration is a precondition for many computer vision applications. Calibration errors, such as wrong model assumptions or imprecise parameter estimation, can deteriorate a system's overall performance, making the reliable detection and quantification of these errors critical. In this work, we introduce an evaluation scheme to capture the fundamental error sources in camera calibration: systematic errors (biases) and uncertainty (variance). The proposed bias detection method uncovers smallest systematic errors and thereby reveals imperfections of the calibration setup and provides the basis for camera model selection. A novel resampling-based uncertainty estimator enables uncertainty estimation under non-ideal conditions and thereby extends the classical covariance estimator. Furthermore, we derive a simple uncertainty metric that is independent of the camera model. In combination, the proposed methods can be used to assess the accuracy of individual calibrations, but also to benchmark new calibration algorithms, camera models, or calibration setups. We evaluate the proposed methods with simulations and real cameras.