Self-Improving SLAM in Dynamic Environments: Learning When to Mask

TL;DR

This paper introduces a self-learning approach for visual SLAM in dynamic environments, enabling the system to automatically decide when to mask moving objects to improve localization and mapping accuracy.

Contribution

It presents a novel self-supervised method that learns when to mask objects without prior motion assumptions, using a new dataset and achieving state-of-the-art results.

Findings

Outperforms existing SLAM methods on TUM RGB-D dataset.

Achieves superior results on KITTI and ConsInv datasets.

Learns to mask objects dynamically without prior motion models.

Abstract

Visual SLAM - Simultaneous Localization and Mapping - in dynamic environments typically relies on identifying and masking image features on moving objects to prevent them from negatively affecting performance. Current approaches are suboptimal: they either fail to mask objects when needed or, on the contrary, mask objects needlessly. Thus, we propose a novel SLAM that learns when masking objects improves its performance in dynamic scenarios. Given a method to segment objects and a SLAM, we give the latter the ability of Temporal Masking, i.e., to infer when certain classes of objects should be masked to maximize any given SLAM metric. We do not make any priors on motion: our method learns to mask moving objects by itself. To prevent high annotations costs, we created an automatic annotation method for self-supervised training. We constructed a new dataset, named ConsInv, which includes challenging real-world dynamic sequences respectively indoors and outdoors. Our method reaches the state of the art on the TUM RGB-D dataset and outperforms it on KITTI and ConsInv datasets.