Research map / Video / Video segmentation and tracking
Restoration of Non-Rigidly Distorted Underwater Images Using a Combination of Compressive Sensing and Local Polynomial Image Representations
Jerin Geo James, Pranay Agrawal, Ajit Rajwade
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 PEOF motion correction after CS (ablation) | CS · Real2 Middle | SSIM: 0.838 -> 0.914 | Improved |
| add PEOF motion correction after CS (ablation) | CS · Real1 Elephant | SSIM: 0.801 -> 0.808 | Improved |
| add PEOF motion correction after CS (ablation) | CS · Synthetic Vision | SSIM: 0.938 -> 0.972 | Improved |
| add CS step before PEOF (ablation) | PEOF · Real1 Elephant | SSIM: 0.763 -> 0.808 | Improved |
| use CS+PEOF instead of SBR-RPCA | restoration method · Real1 Elephant | SSIM: 0.516 -> 0.808 | Improved |
| use CS+PEOF instead of SBR-RPCA | restoration method · Synthetic Vision | RMSE: 0.216 -> 0.066 | Improved |
| use CS+PEOF instead of the single-frame deep learning method | restoration method · Real2 Middle | SSIM: 0.512 -> 0.914 | Improved |
| use CS+PEOF instead of LWB | restoration method · Synthetic Vision | SSIM: 0.446 -> 0.972 | Improved |
| use PEOF instead of CS (ablation) | restoration method · Real1 Elephant | SSIM: 0.801 -> 0.763 | Hurt |
| add RPCA step to remove sparse artifacts (ablation) | CS+PEOF · video restoration | mean-image quality | No clear effect |
| increase number of frames T beyond 100 (ablation) | CS · real sequences | reconstruction quality | No clear effect |
| increase number of tracked salient points N (ablation) | CS · Real2/Real1/Synthetic | reconstruction quality | Improved |
Closest papers by meaning
- Self-MAR: Self-Supervised Multi-Stage Autoregressive Model for Water-Related Optical Image Enhancement (TPAMI 2026 · similarity 0.87)
- Unsupervised Underwater Image Restoration: From a Homology Perspective (AAAI 2022 · similarity 0.87)
- Wavelet-based Fourier Information Interaction with Frequency Diffusion Adjustment for Underwater Image Restoration (CVPR 2024 · similarity 0.86)
- Sea-ing in Low-light (CVPR 2025 · similarity 0.84)
- Looking Into the Water by Unsupervised Learning of the Surface Shape (NeurIPS 2025 · similarity 0.84)
- WaterDiffusion: Learning a Prior-involved Unrolling Diffusion for Joint Underwater Saliency Detection and Visual Restoration (AAAI 2025 · similarity 0.84)
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.
