Research map / 3D vision / Depth and stereo
GroCo: Ground Constraint for Metric Self-Supervised Monocular Depth
Aurélien Cecille, Stefan Duffner, Franck Davoine, Thibault Neveu, Rémi Agier
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 GroCo ground constraint and attention to a ground-prior baseline | Monodepth2-style baseline · KITTI with 15 deg yaw | AbsRel: 0.209 -> 0.136 | Improved |
| add GroCo ground constraint and attention to a ground-prior baseline | Monodepth2-style baseline · KITTI with no rotation | AbsRel: 0.092 -> 0.089 | No clear effect |
| add GroCo ground constraint and attention to a ground-prior baseline | Monodepth2-style baseline · KITTI to DDAD front camera zero-shot | AbsRel: 0.403 -> 0.154 | Improved |
| add GroCo ground constraint and attention to a ground-prior baseline | Monodepth2-style baseline · KITTI to DDAD left camera zero-shot | delta<1.25: 0.798 -> 0.647 | Mixed |
| use GroCo (self-supervised) in place of supervised depth models for cross-camera transfer | DepthFormer and BTS with GeDepth Adaptive · front-trained model on back/left/right cameras | AbsRel mean: 0.62 -> 0.56 | Improved |
| use GroCo instead of supervised GeDepth-Adaptive on KITTI to DDAD | GeDepth-Adaptive · KITTI to DDAD | AbsRel: AbsRel 0.261 -> 0.424; RMSE 16.132 -> 15.366 | Mixed |
| use GroCo scale recovery instead of VADepth | KITTI self-supervised metric depth · KITTI Eigen (improved labels) | AbsRel: 0.120 -> 0.113 | Improved |
Closest papers by meaning
- Digging Into Self-Supervised Monocular Depth Estimation (ICCV 2019 · similarity 0.92)
- On the Uncertainty of Self-Supervised Monocular Depth Estimation (CVPR 2020 · similarity 0.90)
- From-Ground-To-Objects: Coarse-to-Fine Self-supervised Monocular Depth Estimation of Dynamic Objects with Ground Contact Prior (CVPR 2024 · similarity 0.90)
- Depth2Pose: A Pose-Based Benchmark for Monocular Depth Estimation without Ground-Truth Depth (NeurIPS 2026 · similarity 0.88)
- Depth Pro: Sharp Monocular Metric Depth in Less Than a Second (ICLR 2025 · similarity 0.88)
- A Simple yet Universal Framework for Depth Completion (NeurIPS 2024 · 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.
