Research map / Image generation / Image restoration
Removing Diffraction Image Artifacts in Under-Display Camera via Dynamic Skip Connection Network
Ruicheng Feng, Chongyi Li, Huaijin Chen, Shuai Li, Chen Change Loy, Jinwei Gu
Check 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 |
|---|---|---|---|
| Use DISCNet rather than SFTMD | UDC image restoration · Simulated dataset | PSNR: 42.35 -> 43.27 | Improved |
| Use DISCNet rather than SFTMD | UDC image restoration · Simulated dataset | LPIPS: 0.0123 -> 0.0108 | Improved |
| Train the baseline on various kernels instead of one kernel (ablation) | Baseline UNet variant · Simulated dataset, average over kernels | PSNR: 38.55 -> 40.87 | Improved |
| Add image conditioning (ablation) | Baseline with various kernels · Simulated dataset, average over kernels | PSNR: +0.72 dB | Improved |
| Add PSF conditioning (ablation) | Baseline with various kernels · Simulated dataset, average over kernels | PSNR: +1.27 dB | Improved |
| Combine image and PSF conditions (ablation) | Single-scale dynamic convolution model · Simulated dataset, average over kernels | PSNR: 1.18/0.63 dB increase on PSF/image conditions | Improved |
| Use multi-scale rather than single-scale dynamic filter generators (ablation) | DISCNet · Simulated dataset, average over kernels | PSNR: +0.5 dB | Improved |
| Increase dynamic filter size from 3 to 5 (ablation) | DISCNet · Simulated dataset | PSNR: 42.16 -> 42.77 | Improved |
| Increase dynamic filter size from 5 to 7 (ablation) | DISCNet · Simulated dataset | PSNR: 42.77 -> 42.62 | Hurt |
| Increase dynamic filter size from 5 to 9 (ablation) | DISCNet · Simulated dataset | PSNR: 42.77 -> 42.47 | Hurt |
Closest papers by meaning
- Image Restoration for Under-Display Camera (CVPR 2021 · similarity 0.90)
- Decoupling Degradations with Recurrent Network for Video Restoration in Under-Display Camera (AAAI 2024 · similarity 0.88)
- RIFLE: Removal of Image Flicker-Banding via Latent Diffusion Enhancement (NeurIPS 2026 · similarity 0.82)
- Bricker to BRACE: A Bracket Exposure RAW Dataset and Restoration Model for Flicker-Banding (NeurIPS 2026 · similarity 0.82)
- Edge-SD-SR: Low Latency and Parameter Efficient On-device Super-Resolution with Stable Diffusion via Bidirectional Conditioning (CVPR 2025 · similarity 0.81)
- UltraLED: Learning to See Everything in Ultra-High Dynamic Range Scenes (NeurIPS 2025 · similarity 0.81)
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.
