Research map / 3D vision / Gaussian splatting
DyGS-SLAM: Real-Time Accurate Localization and Gaussian Reconstruction for Dynamic Scenes
Xinggang Hu, Chenyangguang Zhang, Mingyuan Zhao, Yuanze Gui, Xiangkui Zhang, Xiangyang Ji
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 dynamic feature detection (DFD) (ablation) | DyGS-SLAM · TUM and BONN sequences (average) | ATE RMSE (cm): 3.4 -> 2.3 | Improved |
| Add dynamic object box correction (DOBC) (ablation) | DyGS-SLAM · TUM and BONN sequences (average) | ATE RMSE (cm): 2.7 -> 2.3 | Improved |
| Add DFD (ablation) | DyGS-SLAM · BONN mv_box2 | ATE RMSE (cm): 4.7 -> 3.1 | Improved |
| Add DOBC (ablation) | DyGS-SLAM · BONN mv_box1 | ATE RMSE (cm): 3.0 -> 2.1 | Improved |
| Add DOBC (ablation) | DyGS-SLAM · TUM s_xyz | ATE RMSE (cm): 1.2 -> 1.1 | Improved |
| Add adaptive feature densification (AFD) (ablation) | DyGS-SLAM · Qualitative rendering of dynamic scenes | rendering quality | Improved |
| Use DyGS-SLAM instead of DG-SLAM | DG-SLAM · TUM RGB-D (average) | ATE RMSE (cm): 1.9 -> 1.8 | Improved |
| Use DyGS-SLAM instead of Photo-SLAM (static sparse-feature base) | Photo-SLAM · BONN RGB-D (average) | ATE RMSE (cm): 45.2 -> 3.1 | Improved |
| Use DyGS-SLAM instead of RoDyn-SLAM for runtime | RoDyn-SLAM · Frame rate | FPS: 1.2 -> 33.9 | Improved |
| Use DyGS-SLAM instead of Dyna-SLAM | Dyna-SLAM · TUM RGB-D (average) | ATE RMSE (cm): 2.2 -> 1.8 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 4 of 4 on most shared results.
- DG-SLAM — DG-SLAM: Robust Dynamic Gaussian Splatting SLAM with Hybrid Pose Optimization (NeurIPS 2024) · better on 3 of 3 shared results
- Photo-SLAM — Photo-SLAM: Real-time Simultaneous Localization and Photorealistic Mapping for Monocular Stereo and RGB-D Cameras (CVPR 2024) · better on 2 of 2 shared results
- SplaTAM — SplaTAM: Splat Track & Map 3D Gaussians for Dense RGB-D SLAM (CVPR 2024) · better on 1 of 1 shared result
- ESLAM — ESLAM: Efficient Dense SLAM System Based on Hybrid Representation of Signed Distance Fields (CVPR 2023) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- DG-SLAM: Robust Dynamic Gaussian Splatting SLAM with Hybrid Pose Optimization (NeurIPS 2024 · similarity 0.91)
- ProDyG: Progressive Dynamic Scene Reconstruction via Gaussian Splatting from Monocular Videos (NeurIPS 2025 · similarity 0.90)
- WildGS-SLAM: Monocular Gaussian Splatting SLAM in Dynamic Environments (CVPR 2025 · similarity 0.89)
- PLGSLAM: Progressive Neural Scene Represenation with Local to Global Bundle Adjustment (CVPR 2024 · similarity 0.87)
- GS-SLAM: Dense Visual SLAM with 3D Gaussian Splatting (CVPR 2024 · similarity 0.87)
- Photo-SLAM: Real-time Simultaneous Localization and Photorealistic Mapping for Monocular Stereo and RGB-D Cameras (CVPR 2024 · 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.
