Research map / Video / Video segmentation and tracking
Unrolling the Shutter: CNN to Correct Motion Distortions
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 |
|---|---|---|---|
| Add directional row- and column-oriented kernel banks to the square-kernel CNN (ablation) | CNN rolling-shutter correction · Chessboard T-only | PSNR (dB): 37.52 -> 38.01 | Helped |
| Add directional row- and column-oriented kernel banks to the square-kernel CNN (ablation) | CNN rolling-shutter correction · Chessboard R-only | PSNR (dB): 37.77 -> 37.83 | Helped |
| Add directional row- and column-oriented kernel banks to the square-kernel CNN (ablation) | CNN rolling-shutter correction · Chessboard T+R | PSNR (dB): 37.30 -> 37.41 | Helped |
| Add directional row- and column-oriented kernel banks to the square-kernel CNN (ablation) | CNN rolling-shutter correction · Combined building dataset | PSNR (dB): 32.19 -> 32.25 | Helped |
| Add directional row- and column-oriented kernel banks to the square-kernel CNN (ablation) | CNN rolling-shutter correction · Caltech building dataset | PSNR (dB): 32.36 -> 32.50 | Helped |
| Add directional row- and column-oriented kernel banks to the square-kernel CNN (ablation) | CNN rolling-shutter correction · LFW face dataset | PSNR (dB): 34.01 -> 34.14 | Helped |
| Use a single-image CNN instead of reference-assisted video correction | RowColCNN · Chessboard R-only | PSNR (dB): 36.84 -> 37.83 | Helped |
| Use a single-image CNN instead of reference-assisted video correction | RowColCNN · Combined building dataset | PSNR (dB): 32.86 -> 32.25 | Hurt |
| Use a single-image CNN instead of reference-assisted video correction | RowColCNN · Caltech building dataset | PSNR (dB): 32.67 -> 32.50 | Hurt |
| Use a single-image CNN instead of reference-assisted video correction | RowColCNN · LFW face dataset | PSNR (dB): 34.75 -> 34.14 | Hurt |
Closest papers by meaning
- Single Image Rolling Shutter Removal with Diffusion Models (AAAI 2025 · similarity 0.88)
- Rolling Shutter Correction with Intermediate Distortion Flow Estimation (CVPR 2024 · similarity 0.87)
- USB-NeRF: Unrolling Shutter Bundle Adjusted Neural Radiance Fields (ICLR 2024 · similarity 0.86)
- Motion Blur Decomposition with Cross-shutter Guidance (CVPR 2024 · similarity 0.86)
- Order-One Rolling Shutter Cameras (CVPR 2025 · similarity 0.83)
- EVS-assisted Joint Deblurring Rolling-Shutter Correction and Video Frame Interpolation through Sensor Inverse Modeling (CVPR 2024 · similarity 0.82)
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.