Research map / 3D vision / Depth and stereo
iMAP: Implicit Mapping and Positioning in Real-Time
Edgar Sucar, Shikun Liu, Joseph Ortiz, Andrew J. Davison
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 iMAP implicit MLP map instead of TSDF fusion | iMAP · Replica (8 scenes) | completion ratio (<5cm %): 75.09 -> 79.06 | Improved |
| use iMAP implicit MLP map instead of TSDF fusion | iMAP · Replica office-3 | completion ratio (<5cm %): 65.87 -> 77.34 | Improved |
| use iMAP instead of TSDF fusion for accuracy | iMAP · Replica (8 scenes) | accuracy (cm): 3.45 -> 4.43 | Hurt |
| use MLP of width 256 instead of TSDF 256^3 voxel grid | iMAP · memory | memory (MB): 67.10 -> 1.04 | Improved |
| use iMAP instead of ORB-SLAM2/BAD-SLAM for tracking | iMAP · TUM RGB-D | ATE RMSE (cm): 1.6 -> 4.9 | Hurt |
| reduce network width from 256 to 128 (ablation) | iMAP · Replica office-2 | completion ratio (<5cm %): 77.22 -> 75.79 | Hurt |
| increase network width from 256 to 512 (ablation) | iMAP · Replica office-2 | completion ratio (<5cm %): 77.22 -> 76.91 | No clear effect |
| reduce keyframe window from 5 to 3 (ablation) | iMAP · Replica office-2 | completion ratio (<5cm %): 77.22 -> 75.82 | Hurt |
| increase keyframe window from 5 to 10 (ablation) | iMAP · Replica office-2 | completion ratio (<5cm %): 77.22 -> 77.35 | No clear effect |
| increase pixel samples from 200 to 400 (ablation) | iMAP · Replica office-2 | completion ratio (<5cm %): 77.22 -> 77.49 | No clear effect |
| reduce pixel samples from 200 to 100 (ablation) | iMAP · Replica office-2 | completion ratio (<5cm %): 77.22 -> 77.33 | No clear effect |
| raise keyframe selection threshold from 0.65 to 0.85 (ablation) | iMAP · Replica office-2 | completion ratio (<5cm %): 77.22 -> 78.03 | No clear effect |
Later papers that compared with this method
5 later accepted papers report iMAP as a baseline in their results tables; 5 of them report a better number on most of the results they share with it.
- ToF-Splatting: Dense SLAM using Sparse Time-of-Flight Depth and Multi-Frame Integration (ICCV 2025 · better on 1 of 1 shared result)
- NARUTO: Neural Active Reconstruction from Uncertain Target Observations (CVPR 2024 · better on 3 of 3 shared results)
- ESLAM: Efficient Dense SLAM System Based on Hybrid Representation of Signed Distance Fields (CVPR 2023 · better on 11 of 11 shared results)
- vMAP: Vectorised Object Mapping for Neural Field SLAM (CVPR 2023 · better on 6 of 7 shared results)
- NICE-SLAM: Neural Implicit Scalable Encoding for SLAM (CVPR 2022 · better on 1 of 1 shared result)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- BAD-SLAM — BAD SLAM: Bundle Adjusted Direct RGB-D SLAM (CVPR 2019) · better on 0 of 3 shared results
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
- Photo-SLAM: Real-time Simultaneous Localization and Photorealistic Mapping for Monocular Stereo and RGB-D Cameras (CVPR 2024 · similarity 0.88)
- Benchmarking Implicit Neural Representation and Geometric Rendering in Real-Time RGB-D SLAM (CVPR 2024 · similarity 0.87)
- Loopy-SLAM: Dense Neural SLAM with Loop Closures (CVPR 2024 · similarity 0.85)
- SLAM3R: Real-Time Dense Scene Reconstruction from Monocular RGB Videos (CVPR 2025 · similarity 0.84)
- 4DTAM: Non-Rigid Tracking and Mapping via Dynamic Surface Gaussians (CVPR 2025 · similarity 0.84)
- MNE-SLAM: Multi-Agent Neural SLAM for Mobile Robots (CVPR 2025 · similarity 0.84)
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.
