Research map / 3D vision / Depth and stereo
Large Depth Completion Model from Sparse Observations
Zhu Yu, zhengyi zhao, Runmin Zhang, Lingteng Qiu, Kejie Qiu, Yisheng He, Siyu Zhu, Zilong Dong, Si-Yuan Cao, Hui-liang 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 Poisson-based coarse depth instead of global alignment (ablation) | LDCM · Estimated depth average | REL: 0.024 -> 0.022 | Improved |
| use Poisson-based coarse depth instead of LWLR (ablation) | LDCM · Estimated depth average | REL: 0.025 -> 0.022 | Improved |
| add Poisson coarse depth versus sparse input only (ablation) | LDCM · Estimated depth average | REL: 0.029 -> 0.022 | Improved |
| add global alignment to Poisson coarse depth (ablation) | Poisson coarse depth · Coarse depth average | REL: 0.147 -> 0.059 | Improved |
| use Poisson instead of global alignment for coarse map (ablation) | coarse depth · Coarse depth average | REL: 0.087 -> 0.059 | Improved |
| replace SI-Log depth head with point map head (ablation) | LDCM · Depth completion average | REL: 0.026 -> 0.022 | Improved |
| replace depth + ray map with point map (ablation) | LDCM · Depth completion average | REL: 0.026 -> 0.022 | Improved |
| replace depth + ray map with point map (ablation) | LDCM · Point map estimation average | REL^p: 0.067 -> 0.045 | Improved |
| apply Poisson alignment to DepthAnythingV2 | DepthAnythingV2 · Average of 5 datasets | REL: 0.071 -> 0.039 | Improved |
| apply Poisson alignment to VGGT | VGGT · Average of 5 datasets | REL: 0.121 -> 0.047 | Improved |
| apply Poisson alignment to MoGe V1 | MoGe V1 · Average of 5 datasets | REL: 0.142 -> 0.035 | Improved |
| add extra real data (DrivingStereo) to training (ablation) | LDCM · Depth completion average | REL: 0.022 -> 0.022 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- G2-MonoDepth — G2-MonoDepth: A General Framework of Generalized Depth Inference From Monocular RGB+X Data (TPAMI 2023) · better on 2 of 2 shared results
- MoGe V2 — MoGe: Unlocking Accurate Monocular Geometry Estimation for Open-Domain Images with Optimal Training Supervision (CVPR 2025) · better on 1 of 1 shared result
- VGGT — VGGT: Visual Geometry Grounded Transformer (CVPR 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
- Zero-shot Depth Completion via Test-time Alignment with Affine-invariant Depth Prior (AAAI 2025 · similarity 0.90)
- A Simple yet Universal Framework for Depth Completion (NeurIPS 2024 · similarity 0.89)
- Depth Pro: Sharp Monocular Metric Depth in Less Than a Second (ICLR 2025 · similarity 0.89)
- Distilling Monocular Foundation Model for Fine-grained Depth Completion (CVPR 2025 · similarity 0.88)
- PatchFusion: An End-to-End Tile-Based Framework for High-Resolution Monocular Metric Depth Estimation (CVPR 2024 · similarity 0.88)
- Depth2Pose: A Pose-Based Benchmark for Monocular Depth Estimation without Ground-Truth Depth (NeurIPS 2026 · 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.
