Research map / 3D vision / Gaussian splatting
De4D-SLAM: Gradient-Isolated Static-Dynamic Decoupling for Monocular SLAM in Dynamic Environments
Zhicheng Fan, Zitong Wu, Zhaoxing Fan, Xiao Zhang, Biao Hou, Bo REN
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 gradient-isolated decoupling (ablation) | De4D-SLAM · TUM RGB-D dynamic (ablation) | ATE (cm): 1.31 -> 1.27 | Improved |
| add gradient-isolated decoupling (ablation) | De4D-SLAM · TUM RGB-D dynamic (ablation) | PSNR (dB): 23.10 -> 27.59 | Improved |
| add gradient-isolated decoupling (ablation) | De4D-SLAM · TUM RGB-D dynamic (ablation) | LPIPS: 0.242 -> 0.164 | Improved |
| add flow-induced initialization (ablation) | De4D-SLAM · TUM RGB-D dynamic (ablation) | PSNR (dB): 20.33 -> 27.59 | Improved |
| add flow-induced initialization (ablation) | De4D-SLAM · TUM RGB-D dynamic (ablation) | ATE (cm): 1.28 -> 1.27 | Improved |
| add SA-KAN instead of pixel-wise MLP (ablation) | De4D-SLAM · TUM RGB-D dynamic (ablation) | ATE (cm): 1.37 -> 1.27 | Improved |
| add SA-KAN instead of pixel-wise MLP (ablation) | De4D-SLAM · TUM RGB-D dynamic (ablation) | PSNR (dB): 26.73 -> 27.59 | Improved |
| add spatial aggregation to the uncertainty predictor (ablation) | De4D-SLAM · TUM RGB-D dynamic (ablation) | LPIPS: 0.186 -> 0.164 | Improved |
| use SA-KAN instead of SA-MLP (ablation) | De4D-SLAM · Bonn RGB-D (ablation) | ATE (cm): 2.04 -> 1.95 | Improved |
| use SA-MLP instead of SA-KAN (ablation) | De4D-SLAM · Bonn RGB-D (ablation) | PSNR (dB): 26.96 vs 26.70 | Mixed |
| add spatial context (SA-KAN vs plain MLP) (ablation) | De4D-SLAM · Bonn RGB-D (ablation) | LPIPS: 0.210 -> 0.200 | Improved |
| add Metric3D geometric prior (ablation) | De4D-SLAM · Bonn RGB-D (ablation) | ATE (cm): 2.42 -> 1.95 | Improved |
Closest papers by meaning
- 4DTAM: Non-Rigid Tracking and Mapping via Dynamic Surface Gaussians (CVPR 2025 · similarity 0.90)
- 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)
- Instant4D: 4D Gaussian Splatting in Minutes (NeurIPS 2025 · similarity 0.89)
- DG-SLAM: Robust Dynamic Gaussian Splatting SLAM with Hybrid Pose Optimization (NeurIPS 2024 · similarity 0.89)
- Flux4D: Flow-based Unsupervised 4D Reconstruction (NeurIPS 2025 · similarity 0.88)
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.
