Research map / 3D vision / Gaussian splatting
Flash-Mono: Feed-Forward Accelerated Gaussian Splatting Monocular SLAM
Zicheng Zhang, Ke Wu, Xiangting Meng, Keyu Liu, Jieru Zhao, Wenchao Ding
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 backend refinement (10 iterations) to feed-forward Gaussian prediction (ablation) | Flash-Mono · ScanNet ablation | PSNR: 20.14 -> 22.41 | Improved |
| use submap clip length 8 instead of shorter or longer (>16) (ablation) | Flash-Mono · Tracking ablation | ATE RMSE: 0.106 lowest at 8 frames | Improved |
| add hidden-state loop closure vs PnP+RANSAC or no loop closure (ablation) | Flash-Mono · Tracking ablation | ATE RMSE | Improved |
| add adaptive voxelization (ablation) | Flash-Mono · Map compactness | Gaussian count: 1.35M -> 0.56M | Improved |
| add adaptive voxelization (ablation) | Flash-Mono · Map compactness | PSNR: 19.70 -> 19.44 | Hurt |
| use Flash-Mono instead of MASt3R-SLAM | Flash-Mono · ScanNetV1 0054 | ATE RMSE (cm): 13.25 -> 11.69 | Improved |
| use Flash-Mono instead of MASt3R-SLAM | Flash-Mono · BundleFusion apt0 | ATE RMSE (cm): 9.65 -> 11.44 | Hurt |
| use Flash-Mono with 20 iterations instead of 250 for MonoGS and S3PO-GS | Flash-Mono · ScanNetV1 | FPS: 0.69 -> 12.71 | Improved |
| use Flash-Mono instead of S3PO-GS | Flash-Mono · KITTI 08 | ATE RMSE (m): 64.74 -> 45.25 | Improved |
| use Flash-Mono instead of S3PO-GS | Flash-Mono · KITTI 08 | PSNR: 17.25 -> 16.12 | Hurt |
Closest papers by meaning
- Gaussian Splatting SLAM (CVPR 2024 · similarity 0.93)
- GS-SLAM: Dense Visual SLAM with 3D Gaussian Splatting (CVPR 2024 · similarity 0.92)
- DG-SLAM: Robust Dynamic Gaussian Splatting SLAM with Hybrid Pose Optimization (NeurIPS 2024 · similarity 0.91)
- SplaTAM: Splat Track & Map 3D Gaussians for Dense RGB-D SLAM (CVPR 2024 · similarity 0.89)
- WildGS-SLAM: Monocular Gaussian Splatting SLAM in Dynamic Environments (CVPR 2025 · similarity 0.89)
- VTGaussian-SLAM: RGBD SLAM for Large Scale Scenes with Splatting View-Tied 3D Gaussians (ICML 2025 · similarity 0.89)
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.
