Research map / Detection and segmentation / 3D detection and driving
Diversity Matters: Fully Exploiting Depth Clues for Reliable Monocular 3D Object Detection
Zhuoling Li, Zhan Qu, Yang Zhou, Jianzhuang Liu, Haoqian Wang, Lihui Jiang
Check 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 |
|---|---|---|---|
| produce diverse depths from direct estimation, height, and keypoint geometry (ablation) | MonoDDE · KITTI validation Car | AP3D70 (%): 18.01 -> 19.75 | Improved |
| use depth from keypoint only instead of direct depth only (ablation) | MonoDDE · KITTI validation Car Moderate | AP3D70 (%): 18.01 -> 18.74 | Improved |
| combine all three depth-solving strategies rather than any two (ablation) | MonoDDE · KITTI validation Car Moderate | AP3D70 (%): 19.18 -> 19.75 | Improved |
| use variance-weighted combination instead of arithmetic mean (ablation) | MonoDDE · KITTI validation Car Moderate | AP3D70 (%): 14.31 -> 19.26 | Improved |
| add iterative outlier selection to weighted depth combination (ablation) | MonoDDE · KITTI validation Car Moderate | AP3D70 (%): 19.26 -> 19.75 | Improved |
| replace iterative outlier selection with minimum-variance selection (ablation) | MonoDDE · KITTI validation Car Moderate | AP3D70 (%): 19.75 -> 19.59 | Hurt |
| use 3D IoU to model conditional confidence instead of no conditional confidence (ablation) | MonoDDE · KITTI validation Car Moderate | AP3D70 (%): 18.15 -> 18.54 | Improved |
| combine depth confidence and 3D box confidence for conditional confidence (ablation) | MonoDDE · KITTI validation Car Moderate | AP3D70 (%): 19.48 -> 19.75 | Improved |
| use diverse depth clues rather than direct depth alone | MonoDDE · KITTI test Car Moderate | AP3D70 (%): 14.17 -> 17.14 | Improved |
| use MonoDDE rather than CaDDN | KITTI test Pedestrian Moderate · KITTI test Pedestrian Moderate | AP3D50 (%): 8.14 -> 7.32 | Hurt |
Closest papers by meaning
- Objects Are Different: Flexible Monocular 3D Object Detection (CVPR 2021 · similarity 0.91)
- DID-M3D: Decoupling Instance Depth for Monocular 3D Object Detection (ECCV 2022 · similarity 0.91)
- MonoCD: Monocular 3D Object Detection with Complementary Depths (CVPR 2024 · similarity 0.91)
- Object-Scene-Camera Decomposition and Recomposition for Data Efficient Monocular 3D Object Detection (IJCV 2026 · similarity 0.90)
- MonoDGP: Monocular 3D Object Detection with Decoupled-Query and Geometry-Error Priors (CVPR 2025 · similarity 0.89)
- MonoMAE: Enhancing Monocular 3D Detection through Depth-Aware Masked Autoencoders (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.
