DPGS-SLAM: Diffusion-Prior and Provenance-Aware Monocular Gaussian Splatting SLAM for Dynamic Visual Sensing
DPGS-SLAM: Diffusion-Prior and Provenance-Aware Monocular Gaussian Splatting SLAM for Dynamic Visual Sensing
This paper presents DPGS-SLAM, a monocular RGB SLAM framework designed for dense mapping in environments with dynamic objects.
The paper introduces DPGS-SLAM, a monocular RGB SLAM framework that combines diffusion-based background completion with provenance-aware Gaussian mapping to handle dynamic objects and occluded static regions, reporting specific accuracy metrics on TUM RGB-D, BONN RGB-D, and StaDy4D benchmarks.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
Reported numbers
average ATE RMSE values
1.2
View original evidence
al RGB observations and predicted depth. Experimental results demonstrated competitive camera tracking, with average ATE RMSE values of 1.2 cm on TUM RGB-D and 2.7 cm on BONN RGB-D. On StaDy4D, DPGS-SLAM improved average PSNR by 4.60 dB over WildGS-SLAM using only monocular RGB input.
Open source S6average ATE RMSE values
2.7
View original evidence
al RGB observations and predicted depth. Experimental results demonstrated competitive camera tracking, with average ATE RMSE values of 1.2 cm on TUM RGB-D and 2.7 cm on BONN RGB-D. On StaDy4D, DPGS-SLAM improved average PSNR by 4.60 dB over WildGS-SLAM using only monocular RGB input.
Open source S6average PSNR improvement
4.6
View original evidence
al RGB observations and predicted depth. Experimental results demonstrated competitive camera tracking, with average ATE RMSE values of 1.2 cm on TUM RGB-D and 2.7 cm on BONN RGB-D. On StaDy4D, DPGS-SLAM improved average PSNR by 4.60 dB over WildGS-SLAM using only monocular RGB input.
Open source S6
- paper: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
al RGB observations and predicted depth. Experimental results demonstrated competitive camera tracking, with average ATE RMSE values of 1.2 cm on TUM RGB-D and 2.7 cm on BONN RGB-D. On StaDy4D, DPGS-SLAM improved average PSNR by 4.60 dB over WildGS-SLAM using only monocular RGB input.
Open source S6
Source：arXiv Robotics — research abstracts · arxiv.org