StereoGeo: an end-to-end stereo camera calibration method

2026-06-12

AI Summary

This work proposes the first end-to-end framework for stereo camera calibration, extending GeoCalib to binocular settings. Unlike conventional methods that rely on calibration patterns or multi-view setups and typically estimate intrinsic or extrinsic parameters in isolation, the proposed approach jointly predicts focal lengths, gravity directions, and relative extrinsics for both left and right cameras using a deep neural network. By integrating feature extraction with a differentiable optimizer, the method eliminates the need for calibration targets or multi-view inputs. Evaluated on real-world benchmark datasets, it achieves competitive accuracy in intrinsic parameter estimation and significantly outperforms existing monocular approaches in extrinsic calibration.

Abstract

In this work, we propose StereoGeo, an end-to-end network-based approach for stereo camera calibration. Our method estimates the focal lengths and gravity directions of the left and right cameras, as well as the relative extrinsic transformation relating them. Existing methods often rely on calibration patterns in structured environments or address only a single camera configuration, being limited to either intrinsic or extrinsic estimation, and depending on a multi-view setups. StereoGeo extends the GeoCalib algorithm, integrating deep neural network feature extraction with a differentiable optimizer. Extensive experiments on real-world benchmarks demonstrate that StereoGeo achieves competitive performance for intrinsic calibration and provides accurate stereo extrinsic estimation, outperforming existing methods that are limited to monocular settings. The dataset used in this work is partially publicly available at https://github.com/meddourimane/StereoGeo-dataset.

Problem

• stereo camera calibration
• intrinsic parameters
• extrinsic parameters
• monocular limitation
• calibration patterns

Innovation

• stereo camera calibration
• end-to-end learning
• differentiable optimization
• deep neural networks
• extrinsic estimation