Research map / Detection and segmentation / 3D detection and driving
Categorical Depth Distribution Network for Monocular 3D Object Detection
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 CaDDN instead of D4LCN | Monocular 3D detection · KITTI test, Car Moderate | AP|R40: +1.69% | Helped |
| Use CaDDN instead of MonoPair | Monocular 3D detection · KITTI test, Pedestrian Moderate | AP|R40: +1.46% | Helped |
| Use CaDDN instead of MonoPSR | Monocular 3D detection · KITTI test, Cyclist Moderate | AP|R40: -1.33% | Hurt |
| Use CaDDN instead of M3D-RPN | Monocular 3D detection · Waymo validation, Vehicle LEVEL_1 (IoU 0.7) | 3D mAP: +4.69% | Helped |
| Add predicted depth distributions to repeated image features (ablation) | CaDDN · KITTI val, Car | AP|R40: +1.50%, +0.77%, +0.46% | Helped |
| Add depth distribution supervision (ablation) | CaDDN · KITTI val, Car | AP|R40: +10.40%, +7.60%, +6.54% | Helped |
| Weight foreground depth-loss pixels separately (α_fg = 3.25, α_bg = 0.25) (ablation) | CaDDN depth loss · KITTI val, Car | AP|R40: +0.67%, +1.07%, +0.91% | Helped |
| Use LID instead of uniform depth discretization (ablation) | CaDDN · KITTI val, Car | AP|R40: +3.17%, +1.21%, +1.09% | Helped |
| Jointly optimize depth estimation and object detection (ablation) | CaDDN · KITTI val, Car | AP|R40: +2.97%, +1.45%, +1.86% | Helped |
| Use the full categorical depth distribution rather than one selected bin (ablation) | CaDDN frustum features · KITTI val, Car | AP|R40: +2.96%, 2.60%, 1.88% | Helped |
| Use SID rather than LID depth discretization (ablation) | CaDDN · KITTI val, Car Moderate | AP|R40: 15.65 vs 16.31 | Hurt |
| Extract image features from Block2 instead of Block1 (ablation) | CaDDN · KITTI val, Car Moderate | AP|R40: 16.31 -> 14.86 | Hurt |
Closest papers by meaning
- Triangulation Learning Network: From Monocular to Stereo 3D Object Detection (CVPR 2019 · similarity 0.90)
- MonoDGP: Monocular 3D Object Detection with Decoupled-Query and Geometry-Error Priors (CVPR 2025 · similarity 0.90)
- MonoCD: Monocular 3D Object Detection with Complementary Depths (CVPR 2024 · similarity 0.89)
- RefineAny3D: Depth Refinement as Semantic Alignment for Monocular 3D Detection (NeurIPS 2026 · similarity 0.88)
- MonoDiff: Monocular 3D Object Detection and Pose Estimation with Diffusion Models (CVPR 2024 · similarity 0.87)
- Geometry-Guided Domain Generalization for Monocular 3D Object Detection (AAAI 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.