Research map / 3D vision / Depth and stereo
Multi-Resolution Monocular Depth Map Fusion by Self-Supervised Gradient-Based Composition
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 |
|---|---|---|---|
| Fuse low- and high-resolution depth using gradient-based composition | LeRes · Middlebury2021 | SqRel: 0.464 -> 0.444 | Helped |
| Fuse low- and high-resolution depth using gradient-based composition | DPT · Multiscopic | SqRel: 4.021 -> 3.998 | Helped |
| Fuse low- and high-resolution depth using gradient-based composition | DPT · Hypersim | SqRel: 0.327 -> 0.311 | Helped |
| Use guided-depth ranking for gradient supervision and low-resolution depth for ILNR (ablation) | Depth fusion network · Middlebury2021 | D³R error: 0.688 -> 0.684 | Helped |
| Use proposed ranking loss instead of SGR ranking on high-resolution depth gradients (ablation) | Depth fusion network · Middlebury2021 | D³R error: 0.722 -> 0.688 | Helped |
| Use guided depth for ILNR and proposed ranking instead of using it only for gradient ranking (ablation) | Depth fusion network · Middlebury2021 | D³R error: 0.684 -> 0.711 | Hurt |
| Add the first convolution layer to extract gradients from high-resolution depth (ablation) | Depth fusion network · Middlebury2021 | SqRel: 0.791 -> 0.468 | Helped |
| Add the first convolution layer to extract gradients from high-resolution depth (ablation) | Depth fusion network · Middlebury2021 | rms: 3.811 -> 2.977 | Helped |
Closest papers by meaning
- HR-Depth: High Resolution Self-Supervised Monocular Depth Estimation (AAAI 2021 · similarity 0.90)
- Boosting Monocular Depth Estimation Models to High-Resolution via Content-Adaptive Multi-Resolution Merging (CVPR 2021 · similarity 0.90)
- Self-Distilled Depth Refinement with Noisy Poisson Fusion (NeurIPS 2024 · similarity 0.89)
- Deep Digging into the Generalization of Self-Supervised Monocular Depth Estimation (AAAI 2023 · similarity 0.89)
- Self-Supervised Monocular Trained Depth Estimation Using Self-Attention and Discrete Disparity Volume (CVPR 2020 · similarity 0.89)
- The Temporal Opportunist: Self-Supervised Multi-Frame Monocular Depth (CVPR 2021 · similarity 0.89)
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.