Research map / 3D vision / Gaussian splatting
GS-LIVM: Real-Time Photo-Realistic LiDAR-Inertial-Visual Mapping with Gaussian Splatting
Yusen Xie, Zhenmin Huang, Jin Wu, Jun Ma
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add structure similarity loss (SS) (ablation) | GS-LIVM · hku campus seq 00 | PSNR: 19.005 -> 19.944 | Improved |
| add delta depth similarity loss (DDS) on top of SS (ns=3) (ablation) | GS-LIVM · hku campus seq 00 | PSNR: 19.944 -> 19.292 | Hurt |
| add structure similarity loss (SS) (ablation) | GS-LIVM · 1018 13 | PSNR: 16.039 -> 15.988 | No clear effect |
| add DDS on top of SS (ns=3) (ablation) | GS-LIVM · 1018 13 | PSNR: 15.988 -> 15.732 | Hurt |
| use ns=4 instead of ns=3 (smaller voxel interval) (ablation) | GS-LIVM · hku campus seq 00 | PSNR: 19.292 -> 20.143 | Improved |
| use ns=4 instead of ns=3 (ablation) | GS-LIVM · 1018 13 | PSNR: 15.732 -> 17.982 | Improved |
| use ns=4 instead of ns=3 (ablation) | GS-LIVM · large scenes (hku main building) | mapping time and GPU memory | Hurt |
| increase Voxel-GPR convergence variance eta from 0.1 to 0.3 (ablation) | GS-LIVM · hku campus seq 00 | mapping time: 1240 s -> 202 s mapping time | Improved |
| increase eta from 0.2 to 0.3 (ablation) | GS-LIVM · 1018 13 | mapping time: 891 s -> 208 s mapping time | Improved |
| add Voxel-GPR based Gaussian initialization instead of raw points or uniform init (ablation) | GS-LIVM · qualitative (Figure 7) | convergence time: 1.3 times longer to converge without | Improved |
Closest papers by meaning
- GS-LiDAR: Generating Realistic LiDAR Point Clouds with Panoramic Gaussian Splatting (ICLR 2025 · similarity 0.89)
- GS-SLAM: Dense Visual SLAM with 3D Gaussian Splatting (CVPR 2024 · similarity 0.87)
- SplatAD: Real-Time Lidar and Camera Rendering with 3D Gaussian Splatting for Autonomous Driving (CVPR 2025 · similarity 0.87)
- LiDAR-RT: Gaussian-based Ray Tracing for Dynamic LiDAR Re-simulation (CVPR 2025 · similarity 0.87)
- Wild-GS: Real-Time Novel View Synthesis from Unconstrained Photo Collections (NeurIPS 2024 · similarity 0.86)
- RPC-GS: Gaussian Splatting with native RPC Rendering for Satellite Imagery (NeurIPS 2026 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
