Research map / Image generation / Image restoration
Conditional Prompt Learning via Degradation Perception for Underwater Image Enhancement
Mingze Yao, Zhiying Jiang, Xianping Fu, Huibing Wang
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 |
|---|---|---|---|
| CPLDP vs best baseline PSNR | UIE methods · UIEB val (full-reference) | PSNR: 22.45 -> 22.60 | Improved |
| CPLDP vs best baseline SSIM | UIE methods · UIEB val (full-reference) | SSIM: 0.914 -> 0.891 | Hurt |
| add both degradation and positive reference prompts (ablation) | CPLDP · UIEB val (full-reference) | PSNR: 19.045 -> 22.597 | Improved |
| add both prompts vs positive reference prompt alone (ablation) | CPLDP · UIEB val (full-reference) | PSNR: 21.107 -> 22.597 | Improved |
| add both prompts vs degradation prompt alone (ablation) | CPLDP · UIEB val (full-reference) | PSNR: 19.582 -> 22.597 | Improved |
| use learned conditional prompts instead of fixed text prompts (ablation) | CPLDP · UIEB val (full-reference) | PSNR: 18.339 -> 22.597 | Improved |
| add dynamic learning (fine-tuning strategy) (ablation) | CPLDP · UIEB val (full-reference) | PSNR (training curve): 30.02 dB -> 31.22dB | Improved |
| add adaptive adjustment loss (ablation) | CPLDP · UIEB val (full-reference) | PSNR (training curve): 30.66dB -> 31.22dB | Improved |
| CPLDP vs best baseline on UCCS CCF | UIE methods · UCCS (non-reference) | CCF: 31.37 -> 30.93 | Hurt |
| CPLDP vs baselines on U45 non-reference metrics | UIE methods · U45 (non-reference) | UCIQE/UIQM/CCF: 0.633 -> 0.636 UCIQE | Improved |
Closest papers by meaning
- Adaptive Dual-domain Learning for Underwater Image Enhancement (AAAI 2025 · similarity 0.88)
- UVEB: A Large-scale Benchmark and Baseline Towards Real-World Underwater Video Enhancement (CVPR 2024 · similarity 0.87)
- NAUTILUS: A Large Multimodal Model for Underwater Scene Understanding (NeurIPS 2025 · similarity 0.86)
- WaterDiffusion: Learning a Prior-involved Unrolling Diffusion for Joint Underwater Saliency Detection and Visual Restoration (AAAI 2025 · similarity 0.86)
- Synergistic Multiscale Detail Refinement via Intrinsic Supervision for Underwater Image Enhancement (AAAI 2024 · similarity 0.85)
- Wavelet-based Fourier Information Interaction with Frequency Diffusion Adjustment for Underwater Image Restoration (CVPR 2024 · 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.
