Research map / 3D vision / Depth and stereo
Mono-SF: Multi-View Geometry Meets Single-View Depth for Monocular Scene Flow Estimation of Dynamic Traffic Scenes
Fabian Brickwedde, Steffen Abraham, Rudolf Mester
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 probabilistic ProbDepthNet depth distributions instead of LRC depth values (ablation) | Mono-SF · KITTI scene flow training set | SF-all error (%): 30.96 -> 21.60 | Improved |
| provide depth as distributions rather than single values (ablation) | Mono-SF · KITTI scene flow training set | SF-all error (%): 33.59 -> 21.60 | Improved |
| add CalibNet recalibration to ProbDepthNet (ablation) | Mono-SF · KITTI scene flow training set | SF-all error (%): 26.91 -> 21.60 | Improved |
| add CalibNet recalibration to ProbDepthNet (ablation) | Mono-SF · KITTI scene flow training set | D1-all error (%): 20.32 -> 16.72 | Improved |
| use probabilistic depth instead of constant-Gaussian depth values (ablation) | Mono-SF · KITTI scene flow training set | D1-all error (%): 25.49 -> 16.72 | Improved |
| run Mono-SF optimization after initialization (ablation) | Mono-SF · KITTI scene flow training set | SF-all error (%): 25.92 -> 21.60 | Improved |
| add photometric term alone to initialization (ablation) | Mono-SF · KITTI scene flow training set | SF-all error (%): 25.92 -> 26.11 | No clear effect |
| add photometric term alone to initialization (ablation) | Mono-SF · KITTI scene flow training set | D1-all error (%): 18.72 -> 21.20 | Hurt |
| add photometric term alone to initialization (ablation) | Mono-SF · KITTI scene flow training set | Fl-all error (%): 15.18 -> 13.85 | Improved |
| add single-view depth term to photometric term (ablation) | Mono-SF · KITTI scene flow training set | SF-all error (%): 26.11 -> 23.67 | Improved |
| add smoothness prior to photometric + depth terms (ablation) | Mono-SF · KITTI scene flow training set | SF-all error (%): 23.67 -> 21.60 | Improved |
| add smoothness prior to photometric + depth terms (ablation) | Mono-SF · KITTI scene flow training set | Fl-all error (%): 13.31 -> 11.85 | Improved |
Later papers that compared with this method
5 later accepted papers report Mono-SF as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.
- MonoPLFlowNet: Permutohedral Lattice FlowNet for Real-Scale 3D Scene Flow Estimation with Monocular Images (ECCV 2022 · better on 9 of 9 shared results)
- Self-Supervised Multi-Frame Monocular Scene Flow (CVPR 2021 · better on 0 of 2 shared results)
- Learning To Segment Rigid Motions From Two Frames (CVPR 2021 · better on 0 of 1 shared result)
- Self-Supervised Monocular Scene Flow Estimation (CVPR 2020 · better on 2 of 3 shared results)
- Upgrading Optical Flow to 3D Scene Flow Through Optical Expansion (CVPR 2020 · 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 1 of 1 on most shared results.
- GeoNet — GeoNet: Geometric Neural Network for Joint Depth and Surface Normal Estimation (CVPR 2018) · better on 2 of 2 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
- Self-Supervised Monocular Scene Flow Estimation (CVPR 2020 · similarity 0.91)
- MonoPLFlowNet: Permutohedral Lattice FlowNet for Real-Scale 3D Scene Flow Estimation with Monocular Images (ECCV 2022 · similarity 0.90)
- Exploiting Semantic Information and Deep Matching for Optical Flow (ECCV 2016 · similarity 0.89)
- Dense Monocular Depth Estimation in Complex Dynamic Scenes (CVPR 2016 · similarity 0.88)
- Zero-Shot Monocular Scene Flow Estimation in the Wild (CVPR 2025 · similarity 0.88)
- Deep Scene Flow Learning: From 2D Images to 3D Point Clouds (TPAMI 2023 · similarity 0.88)
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.
