Slam improves camera tracking and 3d scene reconstruction in dynamic environments
SCOUT-SLAM: Structurally-Coupled Dual Uncertainty-Aware 3DGS SLAM in the Wild
Computer Vision and Pattern Recognition
Summary
Tracking a camera and building a 3D scene map at the same time is tricky when things move fast or the environment changes. Existing methods struggle because their ability to estimate uncertainty depends on how good their 3D scene reconstruction is, which creates a cycle of errors. The authors propose SCOUT-SLAM, which estimates two types of uncertainty together from one shared model to break this cycle. This leads to better camera tracking and cleaner 3D scene maps in challenging, real-world conditions.
What this means in practice
- •For augmented reality developers: Create AR apps that maintain accurate camera tracking and stable 3D maps despite fast camera motion or moving objects.
- •For robotics navigation teams: Enable robots to better localize themselves and map complex, dynamic environments by improving uncertainty estimation for both tracking and mapping.
Authors
Kumaran Karthik, Pramat Shastri Jois, Suresh Sundaram
Abstract
Recently, 3D Gaussian Splatting SLAM (3DGS-SLAM) has gained significant momentum in simultaneous localization and 3DGS scene reconstruction. In real-world scenarios with rapid camera motion and cluttered dynamic environments, existing methods rely on the stability of the underlying scene reconstruction to model uncertainty. This leads to a circular dependency between camera tracking accuracy and reconstruction quality: reconstruction instabilities degrade uncertainty modeling, which affects accurate camera tracking and static scene reconstruction. To address this, the paper proposes SCOUT-SLAM, a structurally-coupled dual-uncertainty framework in which both uncertainties are estimated from a shared base network. A low-rank adaptation of this network, trained on multi-view feature consistency, estimates a tracking uncertainty that does not depend solely on the reconstruction quality. A spatially-adaptive prior modulates the network's training objective so that reconstruction instability does not inflate uncertainty on static regions, keeping the shared representation intact for both branches. Evaluations on dynamic benchmarks (TUM RGB-D, Bonn Dynamic, Wild-SLAM MoCap) demonstrate that SCOUT-SLAM achieves state-of-the-art camera tracking accuracy and artifact-free static scene reconstruction.