CAD-SLAM: Consistency-Aware Dynamic SLAM with Dynamic-Static Decoupled Mapping

TL;DR

CAD-SLAM introduces a novel dynamic SLAM framework that detects and models moving objects in real-time, significantly improving localization and mapping accuracy in dynamic environments by decoupling dynamic and static scene elements.

Contribution

The paper presents a consistency-aware dynamic detection method and a dynamic-static decoupled mapping strategy for real-time dynamic scene understanding in SLAM.

Findings

Achieves state-of-the-art localization accuracy in dynamic scenes.

Effectively detects and segments dynamic objects across categories.

Demonstrates robust dynamic mapping on multiple datasets.

Abstract

Recent advances in neural radiation fields (NeRF) and 3D Gaussian-based SLAM have achieved impressive localization accuracy and high-quality dense mapping in static scenes. However, these methods remain challenged in dynamic environments, where moving objects violate the static-world assumption and introduce inconsistent observations that degrade both camera tracking and map reconstruction. This motivates two fundamental problems: robustly identifying dynamic objects and modeling them online. To address these limitations, we propose CAD-SLAM, a Consistency-Aware Dynamic SLAM framework with dynamic-static decoupled mapping. Our key insight is that dynamic objects inherently violate cross-view and cross-time scene consistency. We detect object motion by analyzing geometric and texture discrepancies between historical map renderings and real-world observations. Once a moving object is identified, we perform bidirectional dynamic object tracking (both backward and forward in time) to achieve complete sequence-wise dynamic recognition. Our consistency-aware dynamic detection model achieves category-agnostic, instantaneous dynamic identification, which effectively mitigates motion-induced interference during localization and mapping. In addition, we introduce a dynamic-static decoupled mapping strategy that employs a temporal Gaussian model for online incremental dynamic modeling. Experiments conducted on multiple dynamic datasets demonstrate the flexible and accurate dynamic segmentation capabilities of our method, along with the state-of-the-art performance in both localization and mapping.