Research map / Detection and segmentation / 3D detection and driving
MonoSOWA: Scalable Monocular 3D Object Detector Without Human Annotations
Jan Skvrna, Lukas Neumann
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 Local Object Motion Model (ablation) | MonoSOWA pseudo-labels · KITTI train Easy | AP_3D@0.5: 17.40 -> 29.05 | Improved |
| add Saturated Closeness Criterion (ablation) | MonoSOWA pseudo-labels · KITTI train Easy | AP_3D@0.5: 29.05 -> 32.29 | Improved |
| add SCC without LOMM (ablation) | MonoSOWA pseudo-labels · KITTI train Easy | AP_3D@0.5: 17.40 -> 17.16 | No clear effect |
| add Canonical Object Space when mixing KITTI and KITTI-360 (ablation) | MonoDETR (human labels) · KITTI val Easy | AP_3D@0.3: 69.62 -> 81.25 | Improved |
| naively add KITTI-360 without COS (ablation) | MonoDETR (human labels) · KITTI val Easy | AP_3D@0.3: 79.72 -> 69.62 | Hurt |
| add real LiDAR for pseudo-label generation (ablation) | MonoSOWA · KITTI val Easy | AP_3D@0.5: 51.55 -> 60.04 | Improved |
| pretrain on pseudo-labels then fine-tune on 1% Waymo human labels | MonoDETR · Waymo val | AP_3D@0.5: 13.81 -> 18.43 | Improved |
| pretrain on pseudo-labels then fine-tune on 5% Waymo human labels | MonoDETR · Waymo val | AP_3D@0.5: 18.12 -> 20.68 | Improved |
| pretrain on pseudo-labels, fine-tune with 15% labels | MonoDETR · Waymo val | AP_3D@0.5: 20.60 -> 20.88 | Improved |
| pretrain on KITTI-360 pseudo-labels, 25% KITTI labels vs 100% human only | MonoDETR · KITTI val Easy | AP_3D@0.7: 29.36 -> 32.64 | Improved |
| increase aggregation frames beyond +-30 (ablation) | MonoSOWA pseudo-labels · KITTI train Easy | AP_BEV@0.5: 40.21 -> 39.23 | No clear effect |
| steepness alpha 10 vs 1 (ablation) | MonoSOWA SCC · KITTI train Moderate | AP_BEV@0.5: 28.28 -> 33.10 | Improved |
Later papers that compared with this method
1 later accepted paper reports MonoSOWA as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- PLOT: Pseudo-Labeling via Object Tracking for Monocular 3D Object Detection (ECCV 2026 · better on 6 of 6 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.
- DEVIANT — DEVIANT: Depth EquiVarIAnt NeTwork for Monocular 3D Object Detection (ECCV 2022) · 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
- MonoDiff: Monocular 3D Object Detection and Pose Estimation with Diffusion Models (CVPR 2024 · similarity 0.90)
- Training an Open-Vocabulary Monocular 3D Detection Model without 3D Data (NeurIPS 2024 · similarity 0.90)
- MonoPlace3D: Learning 3D-Aware Object Placement for 3D Monocular Detection (CVPR 2025 · similarity 0.89)
- LabelAny3D: Label Any Object 3D in the Wild (NeurIPS 2025 · similarity 0.89)
- MonoDGP: Monocular 3D Object Detection with Decoupled-Query and Geometry-Error Priors (CVPR 2025 · similarity 0.89)
- UNION: Unsupervised 3D Object Detection using Object Appearance-based Pseudo-Classes (NeurIPS 2024 · 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.
