Research map / 3D vision / Depth and stereo
MD2E: Modeling Depth-to-Edge Cues for Monocular Metric Depth Estimation
Chao Ning, Minghe Shen, Naoto Yokoya
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 |
|---|---|---|---|
| add SQE edge-score depth calibration (ablation) | MD2E · average over six unseen benchmarks | delta1: 0.251 -> 0.604 | Improved |
| add edge supervision (Ledge and LMSE) (ablation) | MD2E · average over six unseen benchmarks | delta1: 0.325 -> 0.604 | Improved |
| add L1 alignment between predicted and GT edge scores (ablation) | MD2E · average over six unseen benchmarks | delta1: 0.572 -> 0.604 | Improved |
| add edge-depth consistency loss LMSE (ablation) | MD2E · average over six unseen benchmarks | A.Rel: 0.240 -> 0.233 | No clear effect |
| use MD2E instead of Metric3D v2 on zero-shot iBIMS-1 | MD2E · iBIMS-1 zero-shot | A.Rel: 0.185 -> 0.087 | Improved |
| use MD2E instead of UniDepthV2 on zero-shot ETH3D | MD2E · ETH3D zero-shot | A.Rel: 0.256 -> 0.277 | Hurt |
| use MD2E instead of Metric3D v2 on DIODE indoor | MD2E · DIODE indoor zero-shot | RMS: A.Rel 0.093 -> 0.089; RMS 0.389 -> 0.400 | Mixed |
| fine-tune MD2E on NYUv2 versus UniDepthV2 | MD2E · NYUv2 in-domain fine-tuned | A.Rel: 0.047 -> 0.044 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Depth Pro — Depth Pro: Sharp Monocular Metric Depth in Less Than a Second (ICLR 2025) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Metric from Human: Zero-shot Monocular Metric Depth Estimation via Test-time Adaptation (NeurIPS 2024 · similarity 0.90)
- Mind The Edge: Refining Depth Edges in Sparsely-Supervised Monocular Depth Estimation (CVPR 2024 · similarity 0.89)
- Depth Pro: Sharp Monocular Metric Depth in Less Than a Second (ICLR 2025 · similarity 0.89)
- Depth2Pose: A Pose-Based Benchmark for Monocular Depth Estimation without Ground-Truth Depth (NeurIPS 2026 · similarity 0.88)
- UniDepth: Universal Monocular Metric Depth Estimation (CVPR 2024 · similarity 0.88)
- SharpDepth: Sharpening Metric Depth Predictions Using Diffusion Distillation (CVPR 2025 · similarity 0.88)
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.
