Research map / Detection and segmentation / 3D detection and driving
RARE: Learn to RAnk and REtrieve for Monocular 3D Object Detection
Hyeonjeong Park, Peixi Xiong, Xiaoqian Ruan, Dian Jia, Pei Yu, Wei Tang
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 ranking-based confidence loss (point-wise + pair-wise) (ablation) | single-point MonoDETR-style baseline · KITTI val Car AP3D (Easy) | AP3D|40: 24.93 -> 26.54 | Improved |
| add learn-to-retrieve query set without ranking loss (ablation) | single-point MonoDETR-style baseline · KITTI val Car AP3D (Easy) | AP3D|40: 24.93 -> 27.81 | Improved |
| combine ranking and retrieval (ablation) | RARE · KITTI val Car AP3D (Moderate) | AP3D|40: 19.04 -> 22.05 | Improved |
| add point-wise loss alone for confidence (ablation) | depth-uncertainty confidence baseline · KITTI val Car confidence | Pearson correlation: 0.540 -> 0.827 | Improved |
| add pair-wise loss to point-wise loss (ablation) | ranking confidence · KITTI val Car confidence | Spearman correlation: 0.650 -> 0.754 | Improved |
| use pair-wise loss only (no point-wise) (ablation) | ranking confidence · KITTI val Car AP3D (Easy) | AP3D|40: 24.93 -> 25.02 | No clear effect |
| naive query-set supervision (all queries vs same ground truth) (ablation) | query set · KITTI val Car AP3D (Easy) | AP3D|40: 24.93 -> 25.91 | Improved |
| use retrieval-aware multi-hypothesis loss instead of naive supervision (ablation) | query set · KITTI val Car AP3D (Easy) | AP3D|40: 25.91 -> 26.44 | Improved |
| add diversity regularization term on selection distribution (ablation) | retrieval learning loss · KITTI val Car AP3D (Easy) | AP3D|40: 26.44 -> 27.81 | Improved |
| replace MonoDGP with RARE on Car test | MonoDGP · KITTI test Car AP3D (Easy) | AP3D|40: 26.35 -> 28.83 | Improved |
| RARE on Pedestrian category versus MonoUNI/CMKD | KITTI test Pedestrian · KITTI test Pedestrian | AP3D|40: Easy 14.85 vs MonoUNI 15.78; Moderate 10.79 vs 10.34 | Mixed |
| use RARE for cross-dataset depth (KITTI-trained, nuScenes eval) | monocular detectors · nuScenes frontal val | depth MAE (m): 1.33 (MonoCD) -> 1.05 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 5 of 5 on most shared results.
- MonoDGP — MonoDGP: Monocular 3D Object Detection with Decoupled-Query and Geometry-Error Priors (CVPR 2025) · better on 4 of 4 shared results
- MonoUNI — MonoUNI: A Unified Vehicle and Infrastructure-side Monocular 3D Object Detection Network with Sufficient Depth Clues (NeurIPS 2023) · better on 3 of 3 shared results
- MonoCD — MonoCD: Monocular 3D Object Detection with Complementary Depths (CVPR 2024) · better on 2 of 2 shared results
- MonoDETR — MonoDETR: Depth-guided Transformer for Monocular 3D Object Detection (ICCV 2023) · better on 2 of 2 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
- MonoRUn: Monocular 3D Object Detection by Reconstruction and Uncertainty Propagation (CVPR 2021 · similarity 0.90)
- MonoDGP: Monocular 3D Object Detection with Decoupled-Query and Geometry-Error Priors (CVPR 2025 · similarity 0.89)
- Monocular 3D Object Detection with Bounding Box Denoising in 3D by Perceiver (ICCV 2023 · similarity 0.88)
- MonoDiff: Monocular 3D Object Detection and Pose Estimation with Diffusion Models (CVPR 2024 · similarity 0.88)
- MonoPlace3D: Learning 3D-Aware Object Placement for 3D Monocular Detection (CVPR 2025 · similarity 0.87)
- MonoMAE: Enhancing Monocular 3D Detection through Depth-Aware Masked Autoencoders (NeurIPS 2024 · similarity 0.87)
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.
