DynaSLAM: Tracking, Mapping and Inpainting in Dynamic Scenes

TL;DR

DynaSLAM enhances visual SLAM by integrating dynamic object detection and background inpainting, enabling robust mapping and tracking in highly dynamic environments for various camera configurations.

Contribution

It introduces a novel SLAM system that combines multi-view geometry and deep learning for dynamic object detection and background inpainting, improving performance in dynamic scenes.

Findings

Outperforms standard SLAM in dynamic scenarios

Accurately detects moving objects using geometry and deep learning

Estimates static scene maps for long-term applications

Abstract

The assumption of scene rigidity is typical in SLAM algorithms. Such a strong assumption limits the use of most visual SLAM systems in populated real-world environments, which are the target of several relevant applications like service robotics or autonomous vehicles. In this paper we present DynaSLAM, a visual SLAM system that, building over ORB-SLAM2 [1], adds the capabilities of dynamic object detection and background inpainting. DynaSLAM is robust in dynamic scenarios for monocular, stereo and RGB-D configurations. We are capable of detecting the moving objects either by multi-view geometry, deep learning or both. Having a static map of the scene allows inpainting the frame background that has been occluded by such dynamic objects. We evaluate our system in public monocular, stereo and RGB-D datasets. We study the impact of several accuracy/speed trade-offs to assess the limits of the proposed methodology. DynaSLAM outperforms the accuracy of standard visual SLAM baselines in highly dynamic scenarios. And it also estimates a map of the static parts of the scene, which is a must for long-term applications in real-world environments.