Rethinking Generic Camera Models for Deep Single Image Camera Calibration to Recover Rotation and Fisheye Distortion

TL;DR

This paper introduces a generic camera model and a learning-based calibration method to improve the accuracy of camera parameter estimation, especially for fisheye images, outperforming existing methods on large datasets.

Contribution

The paper proposes a novel generic camera model with a closed-form projection calculation and a new loss function, enhancing calibration accuracy for various camera distortions.

Findings

Outperforms conventional methods on large-scale datasets

Effective in recovering rotation and fisheye distortion

First analysis of learning-based methods across different projection types

Abstract

Although recent learning-based calibration methods can predict extrinsic and intrinsic camera parameters from a single image, the accuracy of these methods is degraded in fisheye images. This degradation is caused by mismatching between the actual projection and expected projection. To address this problem, we propose a generic camera model that has the potential to address various types of distortion. Our generic camera model is utilized for learning-based methods through a closed-form numerical calculation of the camera projection. Simultaneously to recover rotation and fisheye distortion, we propose a learning-based calibration method that uses the camera model. Furthermore, we propose a loss function that alleviates the bias of the magnitude of errors for four extrinsic and intrinsic camera parameters. Extensive experiments demonstrated that our proposed method outperformed conventional methods on two largescale datasets and images captured by off-the-shelf fisheye cameras. Moreover, we are the first researchers to analyze the performance of learning-based methods using various types of projection for off-the-shelf cameras.