Research map / Video / Video segmentation and tracking
Rolling Shutter Inversion: Bring Rolling Shutter Images to High Framerate Global Shutter Video
Bin Fan, Yuchao Dai, Hongdong Li
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 normalized scanline offset T (ablation) | RSSR network U · Carla-RS with mask (CRM), first scanline | PSNR: 25.43 -> 30.17 | Improved |
| add geometric constraint of Proposition 1 (ablation) | RSSR network U · Carla-RS with mask (CRM), first scanline | PSNR: 25.95 -> 30.17 | Improved |
| add optical flow residual estimation layer (ablation) | RSSR network U · Carla-RS with mask (CRM), first scanline | PSNR: 29.12 -> 30.17 | Improved |
| add smoothness loss L_s (ablation) | RSSR loss · Carla-RS with mask (CRM), first scanline | PSNR: 29.28 -> 30.17 | Improved |
| add reconstruction loss L_r (ablation) | RSSR loss · Carla-RS with mask (CRM), first scanline | PSNR: 29.44 -> 30.17 | Improved |
| add photometric loss L_p (ablation) | RSSR loss · Carla-RS with mask (CRM), first scanline | PSNR: 29.82 -> 30.17 | Improved |
| initialize F from pre-trained PWC-Net instead of scratch (ablation) | RSSR flow network F · Carla-RS with mask (CRM), first scanline | PSNR: 27.37 -> 29.89 | Improved |
| jointly fine-tune pre-trained PWC-Net with whole model (ablation) | RSSR flow network F · Carla-RS with mask (CRM), first scanline | PSNR: 29.89 -> 30.17 | Improved |
| replace PWC-Net with RAFT (ablation) | RSSR flow network F · Carla-RS with mask (CRM), first scanline | PSNR: 30.17 -> 26.75 | Hurt |
| freeze network parameters (ablation) | RSSR with RAFT · Carla-RS with mask (CRM), first scanline | PSNR: 26.75 -> 29.81 | Improved |
| freeze network parameters (ablation) | RSSR with PWC-Net · Carla-RS with mask (CRM), first scanline | PSNR: 30.17 -> 29.44 | Hurt |
| use constant acceleration propagation with Acceleration-Net instead of constant velocity | RSSR · Fastec-RS (FR) | PSNR: 21.26 -> 21.43 | Improved |
Closest papers by meaning
- Inverting a Rolling Shutter Camera: Bring Rolling Shutter Images to High Framerate Global Shutter Video (ICCV 2021 · similarity 0.98)
- Learning Bilateral Cost Volume for Rolling Shutter Temporal Super-Resolution (TPAMI 2024 · similarity 0.93)
- Context-Aware Video Reconstruction for Rolling Shutter Cameras (CVPR 2022 · similarity 0.93)
- Neural Global Shutter: Learn To Restore Video From a Rolling Shutter Camera With Global Reset Feature (CVPR 2022 · similarity 0.92)
- Rolling Shutter Correction with Intermediate Distortion Flow Estimation (CVPR 2024 · similarity 0.89)
- Single Image Rolling Shutter Removal with Diffusion Models (AAAI 2025 · similarity 0.88)
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.
