Research map / Detection and segmentation / 3D detection and driving
Attention-Based Depth Distillation with 3D-Aware Positional Encoding for Monocular 3D Object Detection
Read the paperCheck 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 |
|---|---|---|---|
| Add ADD distillation | MonoDETR · KITTI Car val | AP_3D@IoU=0.7 Mod.: 20.61 -> 21.94 | Helped |
| Add ADD distillation | CaDDN · KITTI Car val | AP_BE V@IoU=0.7 Mod.: 21.53 -> 23.49 | Helped |
| Add ADD distillation | PatchNet · KITTI Car val | AP_3D@IoU=0.7 Mod.: 16.80 -> 16.92 | Helped |
| Add 3D-aware self-attention to feature distillation (ablation) | MonoDETR feature distillation · KITTI Car val | AP_3D@IoU=0.7 Mod.: 19.86 -> 20.90 | Helped |
| Add target-aware cross-attention to response distillation (ablation) | MonoDETR response distillation · KITTI Car val | AP_3D@IoU=0.7 Mod.: 19.97 -> 21.18 | Helped |
| Combine feature and response distillation with both attention modules (ablation) | MonoDETR · KITTI Car val | AP_3D@IoU=0.7 Mod.: 19.93 -> 21.94 | Helped |
| Use GT-depth positional encoding instead of no positional encoding (ablation) | 3D-aware self-attention · KITTI Car val | AP_3D@IoU=0.7 Mod.: 20.37 -> 20.90 | Helped |
| Replace estimated-depth PE with GT-depth PE (ablation) | 3D-aware self-attention · KITTI Car val | AP_3D@IoU=0.7 Mod.: 20.19 -> 20.90 | Helped |
| Set feature-level background loss weight to 0.10 rather than 0.00 (ablation) | Feature distillation · KITTI Car val | AP_3D@IoU=0.7 Mod.: 20.32 -> 20.90 | Helped |
| Use feature-level background loss weight 0.10 rather than 0.20 (ablation) | Feature distillation · KITTI Car val | AP_3D@IoU=0.7 Mod.: 20.45 -> 20.90 | Helped |
| Set response-level background loss weight to 0.50 rather than 1.00 (ablation) | Response distillation · KITTI Car val | AP_3D@IoU=0.7 Mod.: 20.65 -> 21.18 | Helped |
| Apply ADD instead of the strongest no-extra-data baseline | KITTI monocular 3D detection · KITTI Car test | AP_3D Mod.: 16.46 -> 16.81 | Helped |
Closest papers by meaning
- Weakly Supervised Monocular 3D Detection with a Single-View Image (CVPR 2024 · similarity 0.90)
- FD3D: Exploiting Foreground Depth Map for Feature-Supervised Monocular 3D Object Detection (AAAI 2024 · similarity 0.90)
- RefineAny3D: Depth Refinement as Semantic Alignment for Monocular 3D Detection (NeurIPS 2026 · similarity 0.90)
- MonoMAE: Enhancing Monocular 3D Detection through Depth-Aware Masked Autoencoders (NeurIPS 2024 · similarity 0.90)
- MonoCD: Monocular 3D Object Detection with Complementary Depths (CVPR 2024 · similarity 0.90)
- MonoDGP: Monocular 3D Object Detection with Decoupled-Query and Geometry-Error Priors (CVPR 2025 · 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.