Research map / Image generation / Image restoration
UCMNet: Uncertainty-Aware Context Memory Network for Under-Display Camera Image Restoration
Daehyun Kim, Youngmin Kim, Yoon Ju Oh, Tae Hyun Kim
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 uncertainty-driven loss (UDL) (ablation) | UCMNet with PSNR loss · POLED test | PSNR: 33.38 -> 33.59 | Improved |
| replace UDL with high-frequency UDL (ablation) | UCMNet · POLED test | PSNR: 33.59 -> 33.81 | Improved |
| add hv-attention in UPT (ablation) | UCMNet UPT · POLED test | PSNR: 33.59 -> 33.64 | Improved |
| add Memory Bank cross-attention (ablation) | UCMNet with hv-attention · POLED test | PSNR: 33.64 -> 33.69 | No clear effect |
| use Context Bank for cross-attention (ablation) | UCMNet with hv-attention · POLED test | PSNR: 33.64 -> 33.81 | Improved |
| use uncertainty feature instead of router with no prior (ablation) | UCMNet context generation · POLED test | PSNR: 33.48 -> 33.81 | Improved |
| use uncertainty feature instead of input feature for retrieval (ablation) | UCMNet context generation · POLED test | PSNR: 33.63 -> 33.81 | Improved |
| bank size 512 instead of 256 (ablation) | UCMNet · POLED test | PSNR: 33.78 -> 33.60 | Hurt |
| bank size 256 instead of 128 (ablation) | UCMNet · POLED test | PSNR: 33.71 -> 33.78 | Improved |
| UCMNet vs BNUDC | UCMNet · POLED test | PSNR: 33.39 -> 33.81 | Improved |
| UCMNet vs FSI | UCMNet · SYNTH | PSNR: 45.69 -> 46.71 | Improved |
| UCMNet vs DarkIR on low-light enhancement | UCMNet · LOLv2-Real | PSNR: 23.87 -> 23.10 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- BNUDC — BNUDC: A Two-Branched Deep Neural Network for Restoring Images From Under-Display Cameras (CVPR 2022) · better on 9 of 9 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Decoupling Degradations with Recurrent Network for Video Restoration in Under-Display Camera (AAAI 2024 · similarity 0.88)
- From Zero to Detail: Deconstructing Ultra-High-Definition Image Restoration from Progressive Spectral Perspective (CVPR 2025 · similarity 0.85)
- UHD-processer: Unified UHD Image Restoration with Progressive Frequency Learning and Degradation-aware Prompts (CVPR 2025 · similarity 0.84)
- U-Know-DiffPAN: An Uncertainty-aware Knowledge Distillation Diffusion Framework with Details Enhancement for PAN-Sharpening (CVPR 2025 · similarity 0.84)
- Latent Harmony: Synergistic Unified UHD Image Restoration via Latent Space Regularization and Controllable Refinement (NeurIPS 2025 · similarity 0.84)
- Correlation Matching Transformation Transformers for UHD Image Restoration (AAAI 2024 · 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.
