aiqbench analysis
This research advances real-time 3D reconstruction by integrating visual-inertial tracking directly into the Gaussian Splatting pipeline using an incremental training strategy. By leveraging information from SLAM systems alongside efficient Rust-based GPU implementations, the method eliminates heavy preprocessing bottlenecks to enable immediate interaction in robotics and XR environments. The work demonstrates a significant optimization of the rendering workflow through parallel execution with tracking threads, making high-fidelity 3DGS viable for dynamic applications.
Stipple: Real-Time Incremental Gaussian Splatting with Visual-Inertial Tracking
arXiv:2608.00931v1 Announce Type: new Abstract: 3D Gaussian Splatting (3DGS) provides efficient rendering of photo-realistic scenes, but its heavy preprocessing and training steps make it a poor fit for applications that require real-time reconstruction in robotics or XR. This capability is important since it allows immediate feedback and interaction with new environments. Visual-inertial odometry (VIO) and simultaneous localization and mapping (VI-SLAM) systems, on the other hand, specifically target these real-time applications, which makes them a good choice for integration with 3DGS. We propose a new method that tracks and reconstructs simultaneously in real-time by leveraging an efficient visual-inertial tracking system based on Basalt together with a novel incremental method built on top of Brush, an efficient Rust-based GPU-vendor-agnostic implementation of 3D Gaussian Splatting. We show that many of the heavy preprocessing and training steps of 3DGS can be replaced with a more efficient incremental training strategy that has direct access to the information generated by the visual-inertial tracking system. Furthermore, we propose and combine multiple practical improvements to increase the efficiency of the training pipeline and adapt it to run in real-time, parallel to the tracking thread. This work highlights the value of exploiting the complementary nature of SLAM and 3DGS, and how that can lead to promising results for real-time 3D reconstruction.
Abstract:3D Gaussian Splatting (3DGS) provides efficient rendering of photo-realistic scenes, but its heavy preprocessing and training steps make it a poor fit for applications that require real-time reconstruction in robotics or XR. This capability is important since it allows immediate feedback and interaction with new environments. Visual-inertial odometry (VIO) and simultaneous localization and mapping (VI-SLAM) systems, on the other hand, specifically target these real-time applications, which makes them a good choice for integration with 3DGS. We propose a new method that tracks and reconstructs simultaneously in real-time by leveraging an efficient visual-inertial tracking system based on Basalt together with a novel incremental method built on top of Brush, an efficient Rust-based GPU-vendor-agnostic implementation of 3D Gaussian Splatting. We show that many of the heavy preprocessing and training steps of 3DGS can be replaced with a more efficient incremental training strategy that has direct access to the information generated by the visual-inertial tracking system. Furthermore, we propose and combine multiple practical improvements to increase the efficiency of the training pipeline and adapt it to run in real-time, parallel to the tracking thread. This work highlights the value of exploiting the complementary nature of SLAM and 3DGS, and how that can lead to promising results for real-time 3D reconstruction.
Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
arXiv:2608.00931 [cs.RO]
(or arXiv:2608.00931v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2608.00931 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 306 words
Related coverage
Gaussian-LIC2: LiDAR-Inertial-Camera Gaussian Splatting SLAM
arXiv · Aug 6, 2026
2D GauSS-MI: Efficient Active Scene Reconstruction with Balanced Visual and Geometric Quality
arXiv · Sep 21, 2026
Habitat-GS: A High-Fidelity Navigation Simulator with Dynamic Gaussian Splatting
arXiv · Sep 15, 2026
How to Evaluate General-Purpose Robot Policies for Real-World Deployment
NVIDIA · Jul 12, 2026