Dense simultaneous localization and mapping (SLAM) is crucial for numerous robotic and augmented reality applications. However, current methods are often hampered by the non-volumetric or implicit way they represent a scene. This talk introduces SplaTAM, an approach that leverages explicit volumetric representations, i.e., 3D Gaussians, to enable high-fidelity reconstruction from a single unposed RGB-D camera, surpassing the capabilities of existing methods.

SplaTAM employs a simple online tracking and mapping system tailored to the Gaussian representation. It utilizes a silhouette mask to elegantly capture the presence of scene density. This combination enables several benefits over prior volumetric representations, including fast rendering and optimization, determining if areas have been previously mapped, and structured map expansion by adding more Gaussians. Through extensive experiments, this talk will demonstrate SplaTAM’s up to 2X superior performance in camera pose estimation, map construction, and novel-view synthesis over existing methods, paving the way for more immersive high-fidelity SLAM applications.