Panoramic 3D gaussian splatting improves real-time slam tracking
PanoGS-SLAM: Panoramic 3D Gaussian Splatting SLAM
Computer Vision and Pattern Recognition
Summary
SLAM is a technology that helps robots understand where they are and what their surroundings look like in real time. Most existing methods use cameras with narrow views, which can cause problems when the robot moves fast or looks around a lot. The authors created a new system called PanoGS-SLAM that uses 360-degree panoramic images to help the robot track its position more stably and create better maps. Their method also introduces new techniques to keep the map accurate and improve speed. Tests show this system works better than older methods in various environments.
What this means in practice
- •For robotics developers: Build robots that navigate and map complex environments more reliably using panoramic camera data and 3D Gaussian splatting for accurate real-time SLAM under fast motion.
- •For virtual reality engineers: Generate high-quality 3D maps from panoramic images to improve rendering and tracking stability in dynamic VR environments.
Authors
Yongqi Mao, Hao Shi, Yufan Zhang, Zhonghua Yi, Xiangfei Guo, Kaiwei Wang
Abstract
Real-time dense SLAM is a core capability for robotics applications that require robust localization and high- quality mapping in dynamic or fast-changing environments. Recent 3D Gaussian Splatting (3DGS)-based SLAM methods have shown promising performance, but most are designed for narrow-FoV pinhole cameras, where limited angular coverage weakens pose observability and often leads to unstable photo- metric optimization under rapid motion and large viewpoint changes. We present PanoGS-SLAM, the first panoramic dense SLAM system built on 3D Gaussian Splatting. Our method per- forms differentiable rendering and pose optimization directly in the spherical domain, enabling omnidirectional photometric constraints for more stable tracking. To improve geometric consistency and robustness, we introduce (1) a sphere-consistent photometric loss that compensates for the area distortion of equirectangular projection, and (2) a depth-guided Gaussian initialization strategy that stabilizes incremental mapping in newly observed regions. Extensive experiments on both real and synthetic panoramic benchmarks (PALVIO and SynPano) show that PanoGS-SLAM consistently outperforms geometric and GS-based baselines in tracking accuracy and rendering quality, while achieving fast front-end convergence and real-time perfor- mance. In addition, controlled field-of-view experiments reveal a clear monotonic improvement in optimization conditioning and convergence stability as angular coverage increases, high- lighting the fundamental role of sensing geometry in shaping the optimization landscape of differentiable Gaussian-based SLAM. The source code will be made publicly available.