BEV-ODOM: Reducing Scale Drift in Monocular Visual Odometry with BEV Representation

TL;DR

BEV-ODOM introduces a novel Bird's Eye View-based monocular visual odometry framework that effectively reduces scale drift and improves long-term motion estimation accuracy without requiring depth supervision.

Contribution

The paper presents BEV-ODOM, a new MVO approach leveraging BEV representation and a depth-based PV encoder to enhance scale accuracy without complex optimization.

Findings

Reduces scale drift in long-term sequences

Achieves higher accuracy than existing MVO methods

Performs well across multiple datasets

Abstract

Monocular visual odometry (MVO) is vital in autonomous navigation and robotics, providing a cost-effective and flexible motion tracking solution, but the inherent scale ambiguity in monocular setups often leads to cumulative errors over time. In this paper, we present BEV-ODOM, a novel MVO framework leveraging the Bird's Eye View (BEV) Representation to address scale drift. Unlike existing approaches, BEV-ODOM integrates a depth-based perspective-view (PV) to BEV encoder, a correlation feature extraction neck, and a CNN-MLP-based decoder, enabling it to estimate motion across three degrees of freedom without the need for depth supervision or complex optimization techniques. Our framework reduces scale drift in long-term sequences and achieves accurate motion estimation across various datasets, including NCLT, Oxford, and KITTI. The results indicate that BEV-ODOM outperforms current MVO methods, demonstrating reduced scale drift and higher accuracy.