Research map / 3D vision / Depth and stereo
SharinGAN: Combining Synthetic and Real Data for Unsupervised Geometry Estimation
Koutilya PNVR, Hao Zhou, David Jacobs
Check 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 |
|---|---|---|---|
| Map synthetic and real inputs into a task-relevant shared domain with SharinGAN | monocular depth estimation · KITTI eigen test split, 50m cap | Abs Rel: 0.143 -> 0.109 | Improved |
| Map synthetic and real inputs into a task-relevant shared domain with SharinGAN | monocular depth estimation · Make3D test set | Abs Rel: 0.403 -> 0.377 | Improved |
| Add SharinGAN shared-domain mapping | face normal estimation · Photoface test split | normal accuracy < 20°: 43.6% -> 47.88}$\% | Improved |
| Add SharinGAN shared-domain mapping | face normal estimation · Photoface test split | MAE: 25.5$^{\circ}$ -> 24.0$^{\circ}$ | Improved |
| Add SharinGAN shared-domain mapping | face geometry and lighting estimation · MultiPIE | top-1 accuracy: 80.25 -> 81.83 | Improved |
| Add SharinGAN mapping to synthetic and real training images (ablation) | monocular depth estimation · KITTI eigen test split, 50m cap | Abs Rel: 0.137 -> 0.109 | Improved |
| Add reconstruction loss (ablation) | SharinGAN monocular depth estimator · KITTI eigen test split, 50m cap | Abs Rel: 0.1113 -> 0.109 | Improved |
| Add SharinGAN mapping to synthetic and real training images (ablation) | monocular depth estimation · Make3D test set | Abs Rel: 0.476 -> 0.377 | Improved |
| Add reconstruction loss (ablation) | SharinGAN monocular depth estimator · Make3D test set | Abs Rel: 0.401 -> 0.377 | Improved |
Later papers that compared with this method
1 later accepted paper reports SharinGAN as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- SEDiff: Structure Extraction for Domain Adaptive Depth Estimation via Denoising Diffusion Models (ECCV 2024 · better on 2 of 2 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- SfSNet — SfSNet: Learning Shape, Reflectance and Illuminance of Faces `in the Wild' (CVPR 2018) · 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
- Unsupervised Learning of Geometry From Videos With Edge-Aware Depth-Normal Consistency (AAAI 2018 · similarity 0.88)
- DAViD: Data-efficient and Accurate Vision Models from Synthetic Data (ICCV 2025 · similarity 0.87)
- Unsupervised Depth Estimation, 3D Face Rotation and Replacement (NeurIPS 2018 · similarity 0.87)
- Geometry-Aware Symmetric Domain Adaptation for Monocular Depth Estimation (CVPR 2019 · similarity 0.87)
- Domain Decluttering: Simplifying Images to Mitigate Synthetic-Real Domain Shift and Improve Depth Estimation (CVPR 2020 · similarity 0.87)
- Adaptive Surface Normal Constraint for Geometric Estimation From Monocular Images (TPAMI 2024 · 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.
