Research map / 3D vision / Depth and stereo
Multi-Modal Neural Radiance Field for Monocular Dense SLAM with a Light-Weight ToF Sensor
Xinyang Liu, Yijin Li, Yanbin Teng, Hujun Bao, Guofeng Zhang, Yinda Zhang, Zhaopeng Cui
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 L5 temporal filtering to depth prediction input (ablation) | DELTAR depth prediction · hard cases | RMSE: 0.513 -> 0.366 | Improved |
| add L5 temporal filtering to depth prediction input (ablation) | DELTAR depth prediction · normal cases | RMSE: 0.151 -> 0.145 | Improved |
| add temporal filtering to the SLAM system (ablation) | Ours · indoor L5 dataset tracking (avg) | ATE RMSE (m): 0.132 -> 0.119 | Improved |
| add multi-modal feature grid (zone-level supervision with raw L5) (ablation) | Ours · indoor L5 dataset tracking (avg) | ATE RMSE (m): 0.119 -> 0.107 | Improved |
| add pixel-wise predicted depth as additional supervision | Ours · indoor reconstruction | reconstruction quality | Improved |
| use predicted depth as RGB-D input for ORB-SLAM3 | ORB-SLAM3 · indoor L5 dataset | tracking failure ratio / ATE: lost ratio 42.8% -> 28.5% | Mixed |
| use our method vs NICE-SLAM on noisy predicted depth | NICE-SLAM · indoor L5 dataset | F-score: 0.380 vs 0.604 | Hurt |
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.
- NICE-SLAM — NICE-SLAM: Neural Implicit Scalable Encoding for SLAM (CVPR 2022) · better on 5 of 5 shared results
- iMAP — iMAP: Implicit Mapping and Positioning in Real-Time (ICCV 2021) · 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
- ToF-Splatting: Dense SLAM using Sparse Time-of-Flight Depth and Multi-Frame Integration (ICCV 2025 · similarity 0.88)
- TöRF: Time-of-Flight Radiance Fields for Dynamic Scene View Synthesis (NeurIPS 2021 · similarity 0.88)
- DELTAR: Depth Estimation from a Light-Weight ToF Sensor and RGB Image (ECCV 2022 · similarity 0.88)
- RGB Guided ToF Imaging System: A Survey of Deep Learning-Based Methods (IJCV 2024 · similarity 0.86)
- FloatingFusion: Depth from ToF and Image-Stabilized Stereo Cameras (ECCV 2022 · similarity 0.86)
- IBD-SLAM: Learning Image-Based Depth Fusion for Generalizable SLAM (CVPR 2024 · 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.
