On the Two-View Geometry of Unsynchronized Cameras

TL;DR

This paper introduces new algorithms for estimating camera geometry and time shifts simultaneously from multiple unsynchronized video sequences, enabling robust synchronization even with large time offsets.

Contribution

It develops minimal correspondence-based algorithms for fundamental matrix and homography estimation with unknown time shifts, including an iterative method for large time discrepancies.

Findings

Algorithms successfully estimate camera geometry and time shifts.

Methods are robust to large unsynchronization, up to several seconds.

Validated on synthetic and real-world datasets.

Abstract

We present new methods for simultaneously estimating camera geometry and time shift from video sequences from multiple unsynchronized cameras. Algorithms for simultaneous computation of a fundamental matrix or a homography with unknown time shift between images are developed. Our methods use minimal correspondence sets (eight for fundamental matrix and four and a half for homography) and therefore are suitable for robust estimation using RANSAC. Furthermore, we present an iterative algorithm that extends the applicability on sequences which are significantly unsynchronized, finding the correct time shift up to several seconds. We evaluated the methods on synthetic and wide range of real world datasets and the results show a broad applicability to the problem of camera synchronization.