Research map / Image generation / GANs
Adversarial Regularizers in Inverse Problems
Sebastian Lunz, Ozan Öktem, Carola-Bibiane Schönlieb
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 |
|---|---|---|---|
| Replace Total Variation with an unsupervised adversarial regularizer | Denoising · BSDS denoising | PSNR (dB): 26.3 -> 28.2 | Improved |
| Use an unsupervised adversarial regularizer instead of supervised Denoising N.N. | Denoising · BSDS denoising | PSNR (dB): 28.8 -> 28.2 | Hurt |
| Replace Total Variation with an unsupervised adversarial regularizer | CT reconstruction · LIDC CT high noise | PSNR (dB): 27.7 -> 30.5 | Improved |
| Replace Total Variation with an unsupervised adversarial regularizer | CT reconstruction · LIDC CT low noise | PSNR (dB): 30.0 -> 32.5 | Improved |
| Use an adversarial regularizer instead of RED | CT reconstruction · LIDC CT high noise | PSNR (dB): 29.9 -> 30.5 | Improved |
| Use an adversarial regularizer instead of RED | CT reconstruction · LIDC CT low noise | PSNR (dB): 32.8 -> 32.5 | Hurt |
| Use an adversarial regularizer instead of supervised Post-Processing | CT reconstruction · LIDC CT high noise | PSNR (dB): 31.2 -> 30.5 | Hurt |
| Use an adversarial regularizer instead of supervised Post-Processing | CT reconstruction · LIDC CT low noise | PSNR (dB): 33.6 -> 32.5 | Hurt |
| Add partially reconstructed images to the regularizer training samples | Adversarial Regularizer · CT reconstruction | performance | Improved |
Closest papers by meaning
- Total Deep Variation: A Stable Regularization Method for Inverse Problems (TPAMI 2021 · similarity 0.93)
- Weakly Convex Regularisers for Inverse Problems: Convergence of Critical Points and Primal-Dual Optimisation (ICML 2024 · similarity 0.91)
- A Restoration Network as an Implicit Prior (ICLR 2024 · similarity 0.89)
- NPN: Non-Linear Projections of the Null-Space for Imaging Inverse Problems (NeurIPS 2025 · similarity 0.87)
- Solving Inverse Problems with Latent Diffusion Models via Hard Data Consistency (ICLR 2024 · similarity 0.87)
- pcaGAN: Improving Posterior-Sampling cGANs via Principal Component Regularization (NeurIPS 2024 · similarity 0.87)
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.
