Research map / Label-efficient and robust learning / Self-supervised and contrastive
HR-Depth: High Resolution Self-Supervised Monocular Depth Estimation
Read the paperCheck 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 HR-Depth's redesigned dense skip connections and fSE (ablation) | Monodepth2 · KITTI Eigen split, 640×192, monocular | Abs Rel: 0.115 -> 0.109 | Helped |
| Add dense skip connections (ablation) | Monodepth2 · KITTI Eigen split, 640×192, monocular | Abs Rel: 0.115 -> 0.112 | Helped |
| Add fSE without dense skip connections (ablation) | Monodepth2 · KITTI Eigen split, 640×192, monocular | Abs Rel: 0.115 -> 0.114 | Helped |
| Use SE instead of fSE alongside dense skip connections (ablation) | Monodepth2 · KITTI Eigen split, 640×192, monocular | Abs Rel: 0.115 -> 0.110 | Helped |
| Replace SE with fSE alongside dense skip connections (ablation) | Dense skip connection + SE · KITTI Eigen split, 640×192, monocular | Abs Rel: 0.110 -> 0.109 | Helped |
| Add teacher-disparity supervision | Lite-HR-Depth · KITTI Eigen split, 640×192 | Abs Rel: 0.116 -> 0.112 | Helped |
| Add teacher-disparity supervision | Lite-HR-Depth · KITTI Eigen split, 1024×320 | Abs Rel: 0.111 -> 0.105 | Helped |
| Increase input resolution from 640×192 to 1024×320 | Monodepth2 · KITTI Eigen split, monocular | Abs Rel: 0.115 -> 0.115 | No clear effect |
| Increase input resolution from 640×192 to 1024×320 | Lite-HR-Depth · KITTI Eigen split, monocular, without teacher | Abs Rel: 0.116 -> 0.111 | Helped |
| Use Lite-HR-Depth rather than Monodepth2 at 1024×320 without teacher supervision | KITTI Eigen split, monocular · 1024×320 | Abs Rel: 0.115 -> 0.111 | Helped |
| Pretrain on CityScapes and fine-tune on KITTI | HR-Depth · KITTI Eigen split, 1024×320, monocular and stereo | Abs Rel: 0.107 -> 0.104 | Helped |
Closest papers by meaning
- Self-Supervised Monocular Trained Depth Estimation Using Self-Attention and Discrete Disparity Volume (CVPR 2020 · similarity 0.91)
- The Temporal Opportunist: Self-Supervised Multi-Frame Monocular Depth (CVPR 2021 · similarity 0.91)
- Multi-Resolution Monocular Depth Map Fusion by Self-Supervised Gradient-Based Composition (AAAI 2023 · similarity 0.90)
- Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data (CVPR 2024 · similarity 0.88)
- Improved Monocular Depth Prediction Using Distance Transform Over Pre-semantic Contours with Self-supervised Neural Networks (CVPR 2025 · similarity 0.88)
- Mining Supervision for Dynamic Regions in Self-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.