Research map / Label-efficient and robust learning / Self-supervised and contrastive
Exploiting Pseudo Labels in a Self-Supervised Learning Framework for Improved 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 |
|---|---|---|---|
| Replace the Monodepth2 decoder with the proposed decoder (ablation) | Self-supervised depth teacher, ResNet-50 · KITTI Eigen test set | AbsRel: 0.110 -> 0.104 | Helped |
| Replace the Monodepth2 decoder with the proposed decoder (ablation) | Self-supervised depth teacher, ResNet-18 · KITTI Eigen test set | AbsRel: 0.115 -> 0.112 | Helped |
| Train a student on pseudo labels instead of using the self-supervised teacher directly (ablation) | SD-SSMDE, 192 × 640 · KITTI Eigen test set | AbsRel: 0.108 -> 0.100 | Helped |
| Add automatic scale recovery to pseudo labels (ablation) | SD-SSMDE student · KITTI Eigen test set | AbsRel: 0.102 -> 0.103 | No clear effect |
| Add 3D consistency filtering to automatically scaled pseudo labels (ablation) | SD-SSMDE student · KITTI Eigen test set | AbsRel: 0.103 -> 0.100 | Helped |
| Add 3D consistency filtering with threshold 1.0 (ablation) | SD-SSMDE student · KITTI Eigen test set | AbsRel: 0.103 -> 0.100 | Helped |
| Increase the 3D consistency filtering threshold from 1.0 to 1.5 (ablation) | SD-SSMDE student · KITTI Eigen test set | AbsRel: 0.100 -> 0.101 | Hurt |
| Add automatic scale recovery to pseudo labels (ablation) | SD-SSMDE student, ResNet-50 · KITTI Eigen test set with improved ground truth | AbsRel: 0.109 -> 0.076 | Helped |
| Train with the proposed two-stage self-distillation framework | SD-SSMDE, ResNet-50 · KITTI Eigen test set | AbsRel: 0.110 -> 0.100 | Helped |
| Train with the proposed two-stage self-distillation framework | SD-SSMDE, ResNet-18 · KITTI Eigen test set | AbsRel: 0.105 -> 0.106 | Hurt |
| Use SD-SSMDE student with fixed scaling instead of Monodepth2 | ResNet-50 depth model · Cityscapes | AbsRel: 0.129 -> 0.114 | Helped |
Closest papers by meaning
- Mining Supervision for Dynamic Regions in Self-Supervised Monocular Depth Estimation (CVPR 2024 · similarity 0.89)
- Improved Monocular Depth Prediction Using Distance Transform Over Pre-semantic Contours with Self-supervised Neural Networks (CVPR 2025 · similarity 0.88)
- Intrinsic Image Decomposition for Robust Self-supervised Monocular Depth Estimation on Reflective Surfaces (AAAI 2025 · similarity 0.88)
- Self-supervised Monocular Depth Estimation Robust to Reflective Surface Leveraged by Triplet Mining (ICLR 2025 · similarity 0.87)
- From-Ground-To-Objects: Coarse-to-Fine Self-supervised Monocular Depth Estimation of Dynamic Objects with Ground Contact Prior (CVPR 2024 · similarity 0.86)
- SQLdepth: Generalizable Self-Supervised Fine-Structured Monocular Depth Estimation (AAAI 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.