Information-Theoretic Online Multi-Camera Extrinsic Calibration
Artikel i vetenskaplig tidskrift, 2022

Calibration of multi-camera systems is essential for lifelong use of vision-based headsets and autonomous robots. In this work, we present an information-based framework for online extrinsic calibration of multi-camera systems. While previous work largely focuses on either monocular, stereo, or strictly non-overlapping field-of-view (FoV) setups, we allow arbitrary configurations while also exploiting overlapping pairwise FoV when possible. In order to efficiently solve for the extrinsic calibration parameters, which increase linearly with the number of cameras, we propose a novel entropy-based keyframe measure and bound the backend optimization complexity by selecting informative motion segments that minimize the maximum entropy across all extrinsic parameter partitions. We validate the pipeline on three distinct platforms to demonstrate the generality of the method on resolving the extrinsics and performing downstream tasks. Our code is available at https://github.com/edexheim/info_ext_calib.

Motion segmentation

Calibration and Identification

Cameras

Uncertainty

SLAM

Optimization

Current measurement

Entropy

Calibration

Författare

Eric Dexheimer

Carnegie Mellon University (CMU)

Patrick Peluse

Facebook, Inc.

Jianhui Chen

Facebook, Inc.

James Pritts

Digitala bildsystem och bildanalys

Michael Kaess

Carnegie Mellon University (CMU)

IEEE Robotics and Automation Letters

2377-3766 (ISSN)

Vol. In Press

Ämneskategorier

Inbäddad systemteknik

Robotteknik och automation

Datorseende och robotik (autonoma system)

DOI

10.1109/LRA.2022.3145061

Relaterade dataset

URI: https://github.com/edexheim/info_ext_calib

Mer information

Senast uppdaterat

2022-02-10