# Improving ORB-SLAM3 Accuracy in Dynamic Scenes with YOLO11 Segmentation

## TL;DR

This paper improves the accuracy of ORB-SLAM3 in dynamic environments by integrating YOLO11 segmentation to remove moving objects.

## Contribution

A new system, YOLO11-ORB-SLAM3, that enhances SLAM accuracy in dynamic scenes using instance segmentation.

## Key findings

- The system reduces error by 93% on the TUM RGB-D dataset.
- It maintains computational efficiency while improving robustness in real-world robotic applications.

## Abstract

Traditional Visual SLAM systems, like ORB-SLAM3, often lose accuracy in dynamic environments. This work presents YOLO11-ORB-SLAM3, an enhancement to ORB-SLAM3 for dynamic scenarios, which integrates a YOLO11-based instance segmentation module to detect and exclude dynamic features from the tracking process. The system is designed to work with stereo and RGB-D cameras, and its performance was evaluated on challenging dynamic sequences of the public TUM RGB-D dataset, and also through real-world experiments on a mobile robot using a stereo camera to highlight its robustness and viability for real robotic applications. Experimental results demonstrate that the proposed system outperforms the original ORB-SLAM3, reducing the error by 93% in the public TUM dataset while preserving computational efficiency.