Bayesian Scale Estimation for Monocular SLAM Based on Generic Object Detection for Correcting Scale Drift

TL;DR

This paper introduces a Bayesian method that uses deep learning object detection and prior height information to correct scale drift in monocular SLAM, improving metric accuracy of 3D reconstructions.

Contribution

It presents a novel online Bayesian algorithm that integrates object detection and height priors to estimate and correct scale drift in monocular SLAM.

Findings

Outperforms other monocular SLAM methods in relative translational error

Demonstrates effectiveness on the KITTI dataset

Provides a real-time scale correction approach

Abstract

This work proposes a new, online algorithm for estimating the local scale correction to apply to the output of a monocular SLAM system and obtain an as faithful as possible metric reconstruction of the 3D map and of the camera trajectory. Within a Bayesian framework, it integrates observations from a deep-learning based generic object detector and a prior on the evolution of the scale drift. For each observation class, a predefined prior on the heights of the class objects is used. This allows to define the observations likelihood. Due to the scale drift inherent to monocular SLAM systems, we integrate a rough model on the dynamics of scale drift. Quantitative evaluations of the system are presented on the KITTI dataset, and compared with different approaches. The results show a superior performance of our proposal in terms of relative translational error when compared to other monocular systems.