Research map / 3D vision / Depth and stereo
Composite Focus Measure for High Quality Depth Maps
Parikshit Sakurikar, P. J. Narayanan
Read the paperCheck 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 |
|---|---|---|---|
| use weighted composite of top five focus measures instead of single top focus measures | two-stage DfF pipeline · synthetic buddha and medieval light fields | depth PSNR | Improved |
| use consensus-and-low-correlation selection of five FMs instead of unsupervised feature selection (MCFS-5) (ablation) | composite focus measure · synthetic buddha and medieval light fields | depth PSNR | Improved |
| use more than five focus measures in the composite (ablation) | composite focus measure · 150 focal stacks | depth quality | No clear effect |
| use fewer than five focus measures in the composite (ablation) | composite focus measure · 150 focal stacks | depth quality | Hurt |
| categorize focal stacks by scene type when computing FM ranking (ablation) | FM consensus ranking · subsets of 150 focal stacks | top-five FM ranking | No clear effect |
| fit Laplacian distribution to cFM response instead of taking the focal slice peak (ablation) | depth estimation · focal stack with 11 slices | unique depth values: 11 -> 243 | Improved |
| use cFM pipeline instead of the method of Suwajanakorn et al. | depth estimation · Books focal stack | average depth RMSE (inches): 2.66 -> 0.45 | Improved |
Closest papers by meaning
- Noise Robust Depth From Focus Using a Ring Difference Filter (CVPR 2017 · similarity 0.87)
- Robust Shape from Focus via Multiscale Directional Dilated Laplacian and Recurrent Network (IJCV 2026 · similarity 0.86)
- Depth Pro: Sharp Monocular Metric Depth in Less Than a Second (ICLR 2025 · similarity 0.84)
- DualFocus: Depth from Focus with Spatio-Focal Dual Variational Constraints (NeurIPS 2025 · similarity 0.84)
- Depth Prompting for Sensor-Agnostic Depth Estimation (CVPR 2024 · similarity 0.83)
- DepthMaster: Unified Monocular Depth Estimation for Perspective and Panoramic Images (NeurIPS 2026 · 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.
