Research map / Label-efficient and robust learning / Self-supervised and contrastive
On the Uncertainty of Self-Supervised 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 Self-Teaching by training a student to predict the teacher's depth distribution | Monodepth2 with monocular supervision · KITTI Eigen split | Abs Rel: 0.090 -> 0.087 | Helped |
| Add Self-Teaching by training a student to predict the teacher's depth distribution | Monodepth2 with monocular supervision · KITTI Eigen split | AUSE (RMSE): 2.562 -> 2.009 | Helped |
| Add Self-Teaching by training a student to predict the teacher's depth distribution | Monodepth2 with stereo supervision · KITTI Eigen split | Abs Rel: 0.085 -> 0.084 | Helped |
| Use log-likelihood uncertainty instead of Self-Teaching | Monodepth2 with stereo supervision · KITTI Eigen split | AUSE (RMSE): 1.679 -> 0.938 | Helped |
| Add Self-Teaching by training a student to predict the teacher's depth distribution | Monodepth2 with monocular+stereo supervision · KITTI Eigen split | RMSE: 3.739 -> 3.682 | Helped |
| Add Self-Teaching rather than log-likelihood uncertainty | Monodepth2 with monocular+stereo supervision · KITTI Eigen split | AUSE (RMSE): 1.714 -> 1.654 | Helped |
| Combine bootstrapped ensembles with Self-Teaching (ablation) | Monodepth2 with monocular supervision · KITTI Eigen split | AUSE (RMSE): 2.009 -> 1.924 | Helped |
| Use dropout sampling for uncertainty estimation | Monodepth2 with stereo supervision · KITTI Eigen split | AURG (RMSE): -2.169 | Hurt |
| Use dropout sampling for uncertainty estimation | Monodepth2 with monocular supervision · KITTI Eigen split | Abs Rel: 0.090 -> 0.101 | Hurt |
| Use log-likelihood uncertainty with estimated pose rather than Self-Teaching | Monodepth2 with monocular supervision · KITTI Eigen split | AUSE (RMSE): 2.009 -> 2.562 | Hurt |
| Combine empirical estimation with predictive uncertainty estimation (ablation) | Monodepth2 under monocular supervision · KITTI Eigen split | AUSE (RMSE): 2.562 -> 1.924 (Boot+Self); 2.562 -> 2.449 (Boot+Log) | Mixed |
Closest papers by meaning
- Self-Supervised Monocular Trained Depth Estimation Using Self-Attention and Discrete Disparity Volume (CVPR 2020 · similarity 0.93)
- DME: Unveiling the Bias for Better Generalized Monocular Depth Estimation (AAAI 2024 · similarity 0.88)
- Depth2Pose: A Pose-Based Benchmark for Monocular Depth Estimation without Ground-Truth Depth (NeurIPS 2026 · similarity 0.88)
- GeoDepth: From Point-to-Depth to Plane-to-Depth Modeling for Self-Supervised Monocular Depth Estimation (CVPR 2025 · similarity 0.88)
- SQLdepth: Generalizable Self-Supervised Fine-Structured Monocular Depth Estimation (AAAI 2024 · similarity 0.88)
- Mind The Edge: Refining Depth Edges in Sparsely-Supervised Monocular Depth Estimation (CVPR 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.