Research map / 3D vision / Gaussian splatting
I2-SLAM: Inverting Imaging Process for Robust Photorealistic Dense SLAM
Gwangtak Bae, Changwoon Choi, Hyeongjun Heo, Sang Min Kim, Young Min Kim
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 tone mapping (HDR image formation) to the SLAM pipeline (ablation) | I2-SLAM on ScanNet · ScanNet ablation | PSNR: 22.39 -> 25.62 | Improved |
| add trajectory regularization (ablation) | I2-SLAM on ScanNet · ScanNet ablation | PSNR: 24.80 -> 25.62 | Improved |
| add trajectory regularization (ablation) | I2-SLAM on ScanNet · ScanNet ablation | ATE (cm): 2.66 -> 2.56 | Improved |
| add motion-blur-aware rendering (ablation) | I2-SLAM on ScanNet · ScanNet ablation | ATE (cm): 2.71 -> 2.63 | Improved |
| add motion blur modeling without tone mapping (ablation) | I2-SLAM on ScanNet · ScanNet ablation | PSNR: 23.05 -> 22.36 | Hurt |
| attach I2-SLAM to NeRF-SLAM re-implementation | NeRF-SLAM · TUM-RGBD fr3/office | ATE RMSE (cm): 7.13 -> 1.95 | Improved |
| attach I2-SLAM to SplaTAM for tracking | SplaTAM · ScanNet and synthetic RGBD | ATE RMSE (cm): 5.91 -> 4.59 | Mixed |
| attach I2-SLAM to SplaTAM for rendering on scenes with changing appearance | SplaTAM · ScanNet 0785-00 | PSNR: 19.63 -> 26.40 | Improved |
| pre-deblur inputs with NAFNet before NeRF-SLAM instead of modeling blur physically (ablation) | NeRF-SLAM · TUM-RGBD | PSNR: 26.89 -> 26.61 | No clear effect |
| run I2-SLAM with 20% iterations (I2-SLAM-S) versus SplaTAM-S | SplaTAM · ScanNet 0024-01 | PSNR: 21.43 -> 23.92 | Improved |
| increase the number of virtual cameras for motion blur simulation (ablation) | I2-SLAM RGB · Synthetic Italian-flat-1 | PSNR | Improved |
| attach I2-SLAM to SplaTAM | SplaTAM · ScanNet 0736-00 | PSNR: 24.50 -> 24.07 | Hurt |
Later papers that compared with this method
1 later accepted paper reports I2-SLAM as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Unblur-SLAM: Dense Neural SLAM for Blurry Inputs (CVPR 2026 · better on 3 of 3 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- SplaTAM — SplaTAM: Splat Track & Map 3D Gaussians for Dense RGB-D SLAM (CVPR 2024) · better on 3 of 5 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)
- DG-SLAM: Robust Dynamic Gaussian Splatting SLAM with Hybrid Pose Optimization (NeurIPS 2024 · similarity 0.87)
- GS-SLAM: Dense Visual SLAM with 3D Gaussian Splatting (CVPR 2024 · similarity 0.86)
- VTGaussian-SLAM: RGBD SLAM for Large Scale Scenes with Splatting View-Tied 3D Gaussians (ICML 2025 · similarity 0.86)
- IBD-SLAM: Learning Image-Based Depth Fusion for Generalizable SLAM (CVPR 2024 · similarity 0.86)
- Gaussian Splatting SLAM (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.
