Research map / Large language models / AI and society
GSLAM: A General SLAM Framework and Benchmark
Yong Zhao, Shibiao Xu, Shuhui Bu, Hongkai Jiang, Pengcheng Han
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 |
|---|---|---|---|
| use GSLAM header-only BoW instead of DBoW3 for ORB-6 vocabulary training | BoW vocabulary · DroneMap ORB-6 | training time: 911.4 s -> 230.5 s | Improved |
| use GSLAM BoW instead of DBoW3 for loading ORB-6 vocabulary | BoW vocabulary · DroneMap ORB-6 | load time: 1.1 s -> 7.2ms | Improved |
| use GSLAM BoW instead of DBoW3 for memory on ORB-6 | BoW vocabulary · DroneMap ORB-6 | memory: 246.5MB -> 44.4MB | Improved |
| use GSLAM BoW instead of FBoW for saving ORB-6 vocabulary | BoW vocabulary · DroneMap ORB-6 | save time: 20.6ms -> 34.4ms | Hurt |
| use GSLAM BoW instead of FBoW for SIFT-4 training | BoW vocabulary · DroneMap SIFT-4 | training time: 18.7 s -> 23.5 s | Hurt |
| use GSLAM transform library instead of Sophus for SE(3) multiplication | Lie group transforms · 1e6 runs | time (ms): 55.2 -> 28.6 | Improved |
| use GSLAM transform library instead of TooN for SE(3) exponential | Lie group transforms · 1e6 runs | time (ms): 99.2 -> 152.4 | Hurt |
| use GSLAM transform library instead of TooN for SO(3) point transform | Lie group transforms · 1e6 runs | time (ms): 14.5 -> 15.4 | Hurt |
| use SVO instead of ORB-SLAM for accuracy on TUM RGB-D sequence | monocular SLAM · TUM RGB-D nostructure-texture-near-withloop | absolute pose error | Hurt |
Closest papers by meaning
- GS-SLAM: Dense Visual SLAM with 3D Gaussian Splatting (CVPR 2024 · similarity 0.87)
- ESLAM: Efficient Dense SLAM System Based on Hybrid Representation of Signed Distance Fields (CVPR 2023 · similarity 0.84)
- LRSLAM: Low-rank Representation of Signed Distance Fields in Dense Visual SLAM System (ECCV 2024 · similarity 0.84)
- Loopy-SLAM: Dense Neural SLAM with Loop Closures (CVPR 2024 · similarity 0.82)
- DG-SLAM: Robust Dynamic Gaussian Splatting SLAM with Hybrid Pose Optimization (NeurIPS 2024 · similarity 0.82)
- MASt3R-SLAM: Real-Time Dense SLAM with 3D Reconstruction Priors (CVPR 2025 · similarity 0.82)
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.
