Research map / Image generation / Super-resolution
Joint Estimation of Camera Pose, Depth, Deblurring, and Super-Resolution From a Blurred Image Sequence
Haesol Park, Kyoung Mu Lee
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 |
|---|---|---|---|
| add joint super-resolution to the deblurring/depth model (versus bicubic first then proposed w/o SR) (ablation) | Proposed · Synthetic deblurring average | PSNR (dB): 27.58 -> 28.81 | Improved |
| add joint super-resolution (versus proposed w/o SR then bicubic) (ablation) | Proposed · Synthetic deblurring average | PSNR (dB): 25.77 -> 28.81 | Improved |
| add joint super-resolution (ablation) | Proposed · Dolls depth | depth PSNR (dB): 43.50 -> 45.37 | Improved |
| add joint super-resolution (ablation) | Proposed · Reindeer depth | depth PSNR (dB): 37.41 -> 37.99 | Improved |
| add joint super-resolution (ablation) | Proposed · InteriorScene depth | depth PSNR (dB): 30.19 -> 31.28 | Improved |
| add joint super-resolution (ablation) | Proposed · WorkDesk depth | depth PSNR (dB): 39.90 -> 38.13 | Hurt |
| joint estimation instead of independent super-resolution then deblurring | Proposed vs Wang + Kim and Lee · Synthetic deblurring average | PSNR (dB): 27.63 -> 28.81 | Improved |
| apply independent per-frame deblurring before stereo matching versus bicubic input | Baseline depth · Dolls depth | depth PSNR (dB): 41.79 -> 40.51 | Hurt |
| use blur-aware matching of Lee and Lee instead of the baseline | Bic. + Lee and Lee · Dolls depth | depth PSNR (dB): 41.79 -> 19.76 | Hurt |
Closest papers by meaning
- SuperPrimitive: Scene Reconstruction at a Primitive Level (CVPR 2024 · similarity 0.87)
- DegBins: Degradation-Driven Binning for Depth Super-Resolution (NeurIPS 2026 · similarity 0.86)
- Depth2Pose: A Pose-Based Benchmark for Monocular Depth Estimation without Ground-Truth Depth (NeurIPS 2026 · similarity 0.86)
- Video Depth Anything: Consistent Depth Estimation for Super-Long Videos (CVPR 2025 · similarity 0.85)
- GS-Blur: A 3D Scene-Based Dataset for Realistic Image Deblurring (NeurIPS 2024 · similarity 0.85)
- Sequence Matters: Harnessing Video Models in 3D Super-Resolution (AAAI 2025 · similarity 0.85)
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.
