Research map / 3D vision / Depth and stereo
SCE-SLAM: Scale-Consistent Monocular SLAM via Scene Coordinate Embeddings
Yuchen Wu, Jiahe Li, Xiaohan Yu, Lina Yu, Jin Zheng, Xiao Bai
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 basic scene coordinate branch (ablation) | DPVO + DINOv3 · KITTI Odometry | ATE RMSE (m): 45.84 -> 43.62 | Improved |
| add geometry-guided scale propagation (geometry-modulated attention, reference selection, frame aggregation) (ablation) | SCE-SLAM · KITTI Odometry | ATE RMSE (m): 43.62 -> 31.83 | Improved |
| use SuperPoint keypoint sampling instead of random sampling (ablation) | SCE-SLAM · KITTI Odometry | ATE RMSE (m): 31.83 -> 25.79 | Improved |
| add loop closure (ablation) | SCE-SLAM · KITTI Odometry | ATE RMSE (m): 25.79 -> 14.07 | Improved |
| use DINOv3 features (ablation) | DPVO · KITTI Odometry | ATE RMSE (m): 53.61 -> 45.84 | Improved |
| use DINOv3 features (ablation) | DPV-SLAM++ · KITTI Odometry | ATE RMSE (m): 25.75 -> 22.91 | Improved |
| replace DPVO with SCE-SLAM design, both with DINOv3 (ablation) | DPVO + DINOv3 · KITTI Odometry | ATE RMSE (m): 45.84 -> 25.79 | Improved |
| SCE-SLAM vs DPV-SLAM++ both with loop closure and DINOv3 (ablation) | SCE-SLAM · KITTI Odometry | ATE RMSE (m): 22.91 -> 14.07 | Improved |
| use scene coordinate embeddings instead of external metric depth (MegaSaM) | SCE-SLAM · Waymo Open | ATE RMSE (m): 2.776 -> 0.915 | Improved |
| SCE-SLAM vs DPV-SLAM++ with loop closure | SCE-SLAM · Virtual KITTI | ATE RMSE (m): 0.343 -> 0.280 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- DROID-SLAM — DROID-SLAM: Deep Visual SLAM for Monocular, Stereo, and RGB-D Cameras (NeurIPS 2021) · better on 3 of 3 shared results
- MegaSaM — MegaSaM: Accurate, Fast and Robust Structure and Motion from Casual Dynamic Videos (CVPR 2025) · 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
- Revisiting Monocular SLAM with Spatio-Temporal Scene Modeling (CVPR 2026 · similarity 0.91)
- UniSim-SLAM: Feed-Forward SLAM with Unified Sim(3) Optimization (ECCV 2026 · similarity 0.89)
- MASt3R-SLAM: Real-Time Dense SLAM with 3D Reconstruction Priors (CVPR 2025 · similarity 0.87)
- VGGT-SLAM: Dense RGB SLAM Optimized on the SL(4) Manifold (NeurIPS 2025 · similarity 0.87)
- GLACE: Global Local Accelerated Coordinate Encoding (CVPR 2024 · similarity 0.87)
- MegaSaM: Accurate, Fast and Robust Structure and Motion from Casual Dynamic Videos (CVPR 2025 · 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.
