Camera Motion Estimation from RGB-D-Inertial Scene Flow

TL;DR

This paper presents a new method for estimating camera motion by fusing RGB-D images and inertial data through scene flow, improving accuracy in 3D environments.

Contribution

It introduces a novel formulation that integrates visual and inertial data for camera motion estimation, with flexible multi-frame optimization capabilities.

Findings

Fusion of RGB-D and inertial data improves accuracy

Method performs well on synthetic and real datasets

Flexible data utilization enhances robustness

Abstract

In this paper, we introduce a novel formulation for camera motion estimation that integrates RGB-D images and inertial data through scene flow. Our goal is to accurately estimate the camera motion in a rigid 3D environment, along with the state of the inertial measurement unit (IMU). Our proposed method offers the flexibility to operate as a multi-frame optimization or to marginalize older data, thus effectively utilizing past measurements. To assess the performance of our method, we conducted evaluations using both synthetic data from the ICL-NUIM dataset and real data sequences from the OpenLORIS-Scene dataset. Our results show that the fusion of these two sensors enhances the accuracy of camera motion estimation when compared to using only visual data.