Research map / 3D vision / Human pose estimation
Self-Supervised Human Depth Estimation From Monocular Videos
Feitong Tan, Hao Zhu, Zhaopeng Cui, Siyu Zhu, Marc Pollefeys, Ping Tan
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 |
|---|---|---|---|
| Train the detail estimator with YouTube video self-supervision | ReconNet · Unseen in-the-wild YouTube videos | mid- and high-frequency shape detail | Improved |
| Add validation masking for invalid or occluded pixels (ablation) | Self-supervised ReconNet · Tang test set | MAE: 2.828 -> 2.714 | Improved |
| Add validation masking for invalid or occluded pixels (ablation) | Self-supervised ReconNet · Tang test set | accuracy at 4.0cm: 79.46 -> 80.67 | Improved |
| Replace SSIM loss with SSIMcs loss (ablation) | Self-supervised ReconNet with validation mask · Tang test set | MAE: 2.714 -> 2.609 | Improved |
| Replace SSIM loss with SSIMcs loss (ablation) | Self-supervised ReconNet with validation mask · Tang test set | accuracy at 4.0cm: 80.67 -> 82.13 | Improved |
| Use the proposed method with an SMPL base and detail refinement | Human depth estimation · Tang test set | accuracy at 1.0cm: 33.30 vs 31.47 | Mixed |
| Use the proposed method with an SMPL base and detail refinement | Human depth estimation · Tang test set | MAE: 2.735 -> 2.609 | Improved |
| Use the proposed method with an SMPL base and detail refinement | Human depth estimation · HMD test set | MAE (cm): 3.043 -> 2.944 | Improved |
Closest papers by meaning
- Self-Supervised Monocular Trained Depth Estimation Using Self-Attention and Discrete Disparity Volume (CVPR 2020 · similarity 0.89)
- Self-supervised 3D Representation Learning of Dressed Humans from Social Media Videos (TPAMI 2022 · similarity 0.89)
- Mining Supervision for Dynamic Regions in Self-Supervised Monocular Depth Estimation (CVPR 2024 · similarity 0.88)
- The Temporal Opportunist: Self-Supervised Multi-Frame Monocular Depth (CVPR 2021 · similarity 0.88)
- Depth From Videos in the Wild: Unsupervised Monocular Depth Learning From Unknown Cameras (ICCV 2019 · similarity 0.88)
- Video Depth Anything: Consistent Depth Estimation for Super-Long Videos (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.
