Research map / 3D vision / Depth and stereo
Honey, I Shrunk the Arc de Triomphe!
Yuanbo Xiangli, Hanyu Chen, Xueqing Tsang, Noah Snavely
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 |
|---|---|---|---|
| fine-tune MoGe-2 on MetricScenes | MoGe-2 · MetricScenes test set | metric depth Rel^d (w/o GT cam): 37.2 -> 26.5 | Improved |
| fine-tune MoGe-2 on MetricScenes | MoGe-2 · MetricScenes test set | metric depth delta1 (w/o GT cam): 37.7 -> 73.8 | Improved |
| fine-tune MoGe-2 on MetricScenes | MoGe-2 · MetricScenes test set | metric point Rel^p: 11.7 -> 7.59 | Improved |
| fine-tune MoGe-2 on MetricScenes | MoGe-2 · 10 standard benchmarks | metric depth Rel^d (w/o GT cam): 15.7 -> 15.6 | No clear effect |
| fine-tune MoGe-2 on MetricScenes | MoGe-2 · 10 standard benchmarks | metric point Rel^p: 8.19 -> 8.39 | Hurt |
| fine-tune MoGe-2 on MetricScenes | MoGe-2 · 10 standard benchmarks | scale-invariant point Rel^p: 10.8 -> 10.1 | Improved |
| add AerialMegaDepth and MegaScenes to Stereo4D training (ablation) | WildMoGe · MetricScenes test set | metric depth Rel^d (w/o GT cam): 34.0 -> 26.5 | Improved |
| add Stereo4D to AerialMegaDepth + MegaScenes training (ablation) | WildMoGe · MetricScenes test set | metric depth Rel^d (w/o GT cam): 32.7 -> 26.5 | Improved |
| add MegaScenes to AerialMegaDepth training (ablation) | WildMoGe · MetricScenes test set | metric depth Rel^d (w/o GT cam): 41.0 -> 32.7 | Improved |
| add AerialMegaDepth and MegaScenes to Stereo4D training (ablation) | WildMoGe · 10 standard benchmarks (subset ablation) | metric depth Rel^d (w/o GT cam): 19.0 -> 15.6 | Improved |
| use Stereo4D data in training (ablation) | WildMoGe · 10 standard benchmarks (subset ablation) | boundary F1: 14.5 -> 15.6 | Improved |
| use two-stage edge-aware Poisson completion instead of single-stage BG only (ablation) | depth completion · MetricScenes test set depth labels | foreground person height: 3.26m tall person (BG only) | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- MoGe-2 — MoGe-2: Accurate Monocular Geometry with Metric Scale and Sharp Details (NeurIPS 2025) · better on 9 of 10 shared results
- Depth Pro — Depth Pro: Sharp Monocular Metric Depth in Less Than a Second (ICLR 2025) · better on 2 of 2 shared results
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
- DepthMaster: Unified Monocular Depth Estimation for Perspective and Panoramic Images (NeurIPS 2026 · similarity 0.88)
- RSA: Resolving Scale Ambiguities in Monocular Depth Estimators through Language Descriptions (NeurIPS 2024 · similarity 0.88)
- MoGe-2: Accurate Monocular Geometry with Metric Scale and Sharp Details (NeurIPS 2025 · similarity 0.88)
- UniDepth: Universal Monocular Metric Depth Estimation (CVPR 2024 · similarity 0.87)
- Metric from Human: Zero-shot Monocular Metric Depth Estimation via Test-time Adaptation (NeurIPS 2024 · similarity 0.87)
- SurGe: Improved Surface Geometry in Point Maps (NeurIPS 2026 · 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.
