Research map / 3D vision / Depth and stereo
CroMo: Cross-Modal Learning for Monocular Depth Estimation
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 DF sampling to the stereo-only model (ablation) | ResNet50 stereo (S) · CroMo | RMSE: 2.0642 -> 1.8889 | Helped |
| add i-ToF training cues to stereo-only training (ablation) | ResNet50 · CroMo | RMSE: 2.0642 -> 1.8573 | Helped |
| add structured-light training cues to stereo-only training (ablation) | ResNet50 · CroMo | RMSE: 2.0642 -> 1.7510 | Helped |
| add structured-light cues alongside i-ToF (ablation) | ResNet50 ST · CroMo | RMSE: 1.8573 -> 1.6070 | Helped |
| add temporal training cues to STL (ablation) | ResNet50 STL · CroMo | RMSE: 1.6070 -> 1.4889 | Helped |
| use the proposed architecture instead of ResNet50 | Stereo (S) depth estimation · CroMo | RMSE: 2.0642 -> 1.8889 | Helped |
| use Stokes representation instead of raw polarisation measurements (ablation) | p2d / ResNet50 · CroMo | RMSE: 2.0642 -> 2.1291 | Hurt |
| add i-ToF training cues to stereo-only training (ablation) | ResNet50 · Kitchen | RMSE: 1.2930 -> 0.6520 | Helped |
| add structured-light and temporal training cues to ST (ablation) | ResNet50 · Kitchen | RMSE: 0.6520 -> 0.5431 | Helped |
| use direct structured-light supervision instead of STLM (ablation) | Depth estimation · CroMo 0-10m | RMSE: 1.4889 -> 1.4246 | Helped |
| use STLM instead of direct structured-light supervision (ablation) | Depth estimation · CroMo 0-20m | RMSE: 6.2629 -> 2.9512 | Helped |
| convert polarisation input into RGB-like channels (ablation) | Monocular depth estimation · CroMo | RMSE: 1.4 -> 1.53 | Hurt |
Closest papers by meaning
- The Temporal Opportunist: Self-Supervised Multi-Frame Monocular Depth (CVPR 2021 · similarity 0.89)
- P3Depth: Monocular Depth Estimation With a Piecewise Planarity Prior (CVPR 2022 · similarity 0.89)
- Unsupervised Monocular Depth Estimation With Left-Right Consistency (CVPR 2017 · similarity 0.89)
- Connecting the Dots: Learning Representations for Active Monocular Depth Estimation (CVPR 2019 · similarity 0.89)
- Seurat: From Moving Points to Depth (CVPR 2025 · similarity 0.87)
- DepthMaster: Unified Monocular Depth Estimation for Perspective and Panoramic Images (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.