QueensCAMP: an RGB-D dataset for robust Visual SLAM

TL;DR

This paper introduces a new RGB-D dataset with challenging real-world conditions and camera failure simulations to evaluate and improve the robustness of VSLAM systems in robotics.

Contribution

The authors present a novel dataset with real-world challenging scenarios and open-source tools for simulating camera failures, aiding robustness research in VSLAM.

Findings

Traditional VSLAM algorithms degrade under challenging conditions.

Deep learning-based VO also experiences performance drops.

The dataset enables testing and development of more resilient VSLAM systems.

Abstract

Visual Simultaneous Localization and Mapping (VSLAM) is a fundamental technology for robotics applications. While VSLAM research has achieved significant advancements, its robustness under challenging situations, such as poor lighting, dynamic environments, motion blur, and sensor failures, remains a challenging issue. To address these challenges, we introduce a novel RGB-D dataset designed for evaluating the robustness of VSLAM systems. The dataset comprises real-world indoor scenes with dynamic objects, motion blur, and varying illumination, as well as emulated camera failures, including lens dirt, condensation, underexposure, and overexposure. Additionally, we offer open-source scripts for injecting camera failures into any images, enabling further customization by the research community. Our experiments demonstrate that ORB-SLAM2, a traditional VSLAM algorithm, and TartanVO, a Deep Learning-based VO algorithm, can experience performance degradation under these challenging conditions. Therefore, this dataset and the camera failure open-source tools provide a valuable resource for developing more robust VSLAM systems capable of handling real-world challenges.