Research map / Label-efficient and robust learning / Domain adaptation and generalization
Geometry-Aware Symmetric Domain Adaptation 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 |
|---|---|---|---|
| translate synthetic training images to real style (ablation) | synthetic-only depth estimation · KITTI Eigen split | Abs Rel: 0.253 -> 0.229 | Helped |
| train the style translator and depth estimator end-to-end (ablation) | SYN2REAL · KITTI Eigen split | Abs Rel: 0.229 -> 0.220 | Helped |
| add stereo geometry consistency (ablation) | SYN2REAL-E2E · KITTI Eigen split | Abs Rel: 0.220 -> 0.152 | Helped |
| add synthetic labeled data and domain adaptation (ablation) | real-only stereo training · KITTI Eigen split | Abs Rel: 0.158 -> 0.153 | Helped |
| add the reverse real-to-synthetic path and symmetric training (ablation) | SYN2REAL-GC-E2E · KITTI Eigen split | Abs Rel: 0.152 -> 0.160 | Hurt |
| add depth consistency between the two depth predictions (ablation) | GASDA-w/oDC · KITTI Eigen split | Abs Rel: 0.151 -> 0.149 | Helped |
| combine the F_t and F_s predictions at inference (ablation) | GASDA · KITTI Eigen split | Abs Rel: 0.150 -> 0.149; 0.156 -> 0.149 | Helped |
| use vKITTI synthetic supervision with GASDA domain adaptation | KITTI depth estimation · KITTI Eigen split, 80m | Abs Rel: 0.158 -> 0.149 | Helped |
| use GASDA instead of Atapour et al. | KITTI stereo 2015 · 200 training images | Sq Rel: 1.048 -> 0.987 | Helped |
| use GASDA instead of Atapour et al. | Make3D depth estimation without fine-tuning · Make3D test, depth <70m | Abs Rel: 0.423 -> 0.403 | Helped |
Closest papers by meaning
- AdaDepth: Unsupervised Content Congruent Adaptation for Depth Estimation (CVPR 2018 · similarity 0.90)
- Domain Decluttering: Simplifying Images to Mitigate Synthetic-Real Domain Shift and Improve Depth Estimation (CVPR 2020 · similarity 0.89)
- Real-Time Monocular Depth Estimation Using Synthetic Data With Domain Adaptation via Image Style Transfer (CVPR 2018 · similarity 0.89)
- DepthFM: Fast Generative Monocular Depth Estimation with Flow Matching (AAAI 2025 · similarity 0.86)
- Distilling Monocular Foundation Model for Fine-grained Depth Completion (CVPR 2025 · similarity 0.86)
- SharpDepth: Sharpening Metric Depth Predictions Using Diffusion Distillation (CVPR 2025 · similarity 0.86)
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.