Research map / Detection and segmentation / 3D detection and driving
Task-Aware Monocular Depth Estimation for 3D Object Detection
Xinlong Wang, Wei Yin, Tao Kong, Yuning Jiang, Lei Li, Chunhua Shen
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 |
|---|---|---|---|
| Use ForeSeE instead of DenseDepth | KOD depth estimation · KOD foreground | absRel: 0.135 -> 0.118 | Improved |
| Use ForeSeE instead of DenseDepth | KOD depth estimation · KOD global | absRel: 0.138 -> 0.138 | No clear effect |
| Use ForeSeE-generated pseudo-LiDAR instead of baseline pseudo-LiDAR | AVOD · KITTI car easy | AP3D: 7.5 -> 15.0 | Improved |
| Use ForeSeE-generated pseudo-LiDAR instead of baseline pseudo-LiDAR | AVOD · KITTI car moderate | AP3D: 6.1 -> 12.5 | Improved |
| Use ForeSeE-generated pseudo-LiDAR instead of baseline pseudo-LiDAR | AVOD · KITTI car hard | AP3D: 5.4 -> 12.0 | Improved |
| Add foreground-background sensitive loss (ablation) | ForeSeE with separate decoders and objectives · KOD foreground | absRel: 0.120 -> 0.118 | Improved |
| Add foreground-background sensitive loss (ablation) | ForeSeE with separate decoders and objectives · KOD background | absRel: 0.141 -> 0.141 | No clear effect |
| Add separate decoders while retaining separate objectives (ablation) | ForeSeE · KOD background | absRel: 0.147 -> 0.141 | Improved |
| Add separate decoders while retaining separate objectives (ablation) | ForeSeE · KOD foreground | absRel: 0.120 -> 0.120 | No clear effect |
| Use separate foreground and background objectives instead of one shared objective (ablation) | Baseline depth estimator · KOD foreground | absRel: 0.129 -> 0.120 | Improved |
| Use separate foreground and background objectives instead of one shared objective (ablation) | Baseline depth estimator · KOD background | absRel: 0.143 -> 0.147 | Hurt |
| Replace mask-based merging with mask-free element-wise maximum merging (ablation) | ForeSeE inference · KOD | absRel: 0.117 -> 0.118 | Hurt |
Closest papers by meaning
- MonoCD: Monocular 3D Object Detection with Complementary Depths (CVPR 2024 · similarity 0.89)
- MonoGround: Detecting Monocular 3D Objects From the Ground (CVPR 2022 · similarity 0.89)
- FD3D: Exploiting Foreground Depth Map for Feature-Supervised Monocular 3D Object Detection (AAAI 2024 · similarity 0.89)
- MonoDGP: Monocular 3D Object Detection with Decoupled-Query and Geometry-Error Priors (CVPR 2025 · similarity 0.89)
- MonoDiff: Monocular 3D Object Detection and Pose Estimation with Diffusion Models (CVPR 2024 · 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.
