Research map / 3D vision / Depth and stereo
Robust Light Field Depth Estimation for Noisy Scene With Occlusion
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 |
|---|---|---|---|
| combine angular entropy and adaptive defocus costs, then optimize (ablation) | Proposed light-field depth estimation · Buddha, Noisy Buddha, StillLife, Noisy StillLife | mean squared error: 0.0058 -> 0.0047; 0.1174 -> 0.0989; 0.0176 -> 0.0206; 0.0608 -> 0.0532 | Mixed |
| combine angular entropy and adaptive defocus costs (ablation) | Proposed light-field depth estimation · Mona, σ = 0, 5, 10, 15 | mean squared error: 0.0118 -> 0.0081; 0.0381 -> 0.0337; 0.1152 -> 0.1011; 0.2043 -> 0.1792 | Helped |
| add adaptive defocus cost to correspondence cost (ablation) | Proposed light-field depth estimation · Buddha, Noisy Buddha, StillLife, Noisy StillLife | mean squared error: 0.0058 -> 0.0047; 0.1174 -> 0.0989; 0.0176 -> 0.0206; 0.0608 -> 0.0532 | Mixed |
| add angular entropy cost to adaptive defocus cost (ablation) | Proposed light-field depth estimation · Mona, σ = 0, 5, 10, 15 | mean squared error: 0.0217 -> 0.0081; 0.0954 -> 0.0337; 0.1962 -> 0.1011; 0.2788 -> 0.1792 | Helped |
| use the proposed method with optimization instead of the strongest listed optimized baseline | Light-field depth estimation · Buddha, Noisy Buddha, Noisy StillLife | mean squared error: 0.0041 -> 0.0036; 0.0250 -> 0.0170; 0.0432 -> 0.0251 | Helped |
| use the proposed method with optimization instead of the strongest listed optimized baseline | Light-field depth estimation · StillLife | mean squared error: 0.0126 -> 0.0142 | Hurt |
| use the proposed method with optimization instead of the strongest listed optimized baseline | Light-field depth estimation · Mona, σ = 0, 5, 10, 15 | mean squared error: 0.0052 -> 0.0045; 0.0179 -> 0.0072; 0.0224 -> 0.0125; 0.0226 -> 0.0185 | Helped |
Closest papers by meaning
- Trust Your Model: Light Field Depth Estimation With Inline Occlusion Handling (CVPR 2018 · similarity 0.88)
- Accurate Depth and Normal Maps From Occlusion-Aware Focal Stack Symmetry (CVPR 2017 · similarity 0.87)
- Occlusion-Aware Cost Constructor for Light Field Depth Estimation (CVPR 2022 · similarity 0.86)
- What Sparse Light Field Coding Reveals About Scene Structure (CVPR 2016 · similarity 0.86)
- Towards Multimodal Depth Estimation From Light Fields (CVPR 2022 · similarity 0.85)
- Blurry-Edges: Photon-Limited Depth Estimation from Defocused Boundaries (CVPR 2025 · similarity 0.83)
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.