Research map / Point clouds / Segmentation and detection
Learning Depth-Guided Convolutions 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 D⁴LCN instead of AM3D | KITTI monocular Car detection · KITTI test, Car, Moderate | AP|R40: 10.74 -> 11.72 | Helped |
| Add Corner Loss to 3DNet (ablation) | 3DNet on KITTI · KITTI split1, Car, Moderate | AP|R11: 14.61 -> 15.57 | Helped |
| Add depth-guided depth-wise local filtering (ablation) | 3DNet + Corner Loss · KITTI split1, Car, Moderate | AP|R11: 15.57 -> 17.92 | Helped |
| Add shift-pooling to DLCN (ablation) | DLCN · KITTI split1, Car, Moderate | AP|R11: 17.92 -> 19.02 | Helped |
| Add adaptive dilation rates to SP-DLCN (ablation) | SP-DLCN · KITTI split1, Car, Moderate | AP|R11: 19.02 -> 21.71 | Helped |
| Use a depth-wise local convolution without shift-pooling, rather than depth-guided local convolution (ablation) | Depth-guided local convolution · KITTI split1, Car, Moderate | AP|R11: 18.42 -> 17.92 | Hurt |
| Set the shift-pooling channel count to 3 (ablation) | D⁴LCN · KITTI split1, Car 3D detection | AP|R11 | Helped |
| Use progressively more accurate depth maps (ablation) | D⁴LCN · KITTI split1, Car, Moderate | AP|R11: 19.63 -> 21.71 -> 24.06 -> 25.41 | Helped |
| Use D⁴LCN instead of dynamic convolution | KITTI split1, Car 3D detection · KITTI split1, Car, Moderate | AP|R11: 17.67 -> 21.71 | Helped |
| Use D⁴LCN instead of dynamic local filtering | KITTI split1, Car 3D detection · KITTI split1, Car, Moderate | AP|R11: 18.42 -> 21.71 | Helped |
| Use D⁴LCN instead of deformable convolution | KITTI split1, Car 3D detection · KITTI split1, Car, Moderate | AP|R11: 18.24 -> 21.71 | Helped |
Closest papers by meaning
- Multi-Level Fusion Based 3D Object Detection From Monocular Images (CVPR 2018 · similarity 0.90)
- Attention-Based Depth Distillation with 3D-Aware Positional Encoding for Monocular 3D Object Detection (AAAI 2023 · similarity 0.90)
- MonoMAE: Enhancing Monocular 3D Detection through Depth-Aware Masked Autoencoders (NeurIPS 2024 · similarity 0.88)
- MonoCD: Monocular 3D Object Detection with Complementary Depths (CVPR 2024 · similarity 0.88)
- MonoDGP: Monocular 3D Object Detection with Decoupled-Query and Geometry-Error Priors (CVPR 2025 · similarity 0.87)
- FD3D: Exploiting Foreground Depth Map for Feature-Supervised 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.