NGD-SLAM: Towards Real-Time Dynamic SLAM without GPU

TL;DR

This paper presents NGD-SLAM, a real-time dynamic SLAM system that operates solely on a CPU, using a novel mask propagation and hybrid tracking approach to achieve high accuracy and 60 FPS without GPU reliance.

Contribution

It introduces a CPU-only dynamic SLAM system with mask propagation and hybrid tracking, enabling real-time performance without GPU dependency.

Findings

Achieves 60 FPS tracking on a laptop CPU.

Maintains high localization accuracy in dynamic environments.

Operates effectively without GPU support.

Abstract

Many existing visual SLAM methods can achieve high localization accuracy in dynamic environments by leveraging deep learning to mask moving objects. However, these methods incur significant computational overhead as the camera tracking needs to wait for the deep neural network to generate mask at each frame, and they typically require GPUs for real-time operation, which restricts their practicality in real-world robotic applications. Therefore, this paper proposes a real-time dynamic SLAM system that runs exclusively on a CPU. Our approach incorporates a mask propagation mechanism that decouples camera tracking and deep learning-based masking for each frame. We also introduce a hybrid tracking strategy that integrates ORB features with optical flow methods, enhancing both robustness and efficiency by selectively allocating computational resources to input frames. Compared to previous methods, our system maintains high localization accuracy in dynamic environments while achieving a tracking frame rate of 60 FPS on a laptop CPU. These results demonstrate the feasibility of utilizing deep learning for dynamic SLAM without GPU support. Since most existing dynamic SLAM systems are not open-source, we make our code publicly available at: https://github.com/yuhaozhang7/NGD-SLAM