AI Paper Atlas

Research map / Probabilistic and scientific modeling / Variational and generative sampling

Total Deep Variation for Linear Inverse Problems

Erich Kobler, Alexander Effland, Karl Kunisch, Thomas Pock

CVPR 2020 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Posterior sampling with diffusion guidance · 81 citations (OpenAlex)

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 changedWhereMeasuredEffect
Individually train TDV³ for each noise level instead of using the σ=25 model with rescalingTDV denoising · Set12, BSDS68, and Urban100; σ=15, 25, 50PSNR: 33.01 vs 32.93; 27.59 vs 27.50Improved
Use TDV³ instead of FOCNetDenoising · Set12; σ=15, 25, 50PSNR: 33.07 vs 33.01; 30.73 vs 30.66; 27.68 vs 27.59Hurt
Use TDV³ instead of FOCNetDenoising · BSDS68; σ=15, 25, 50PSNR: 31.83 vs 31.82; 29.38 vs 29.37; 26.50 vs 26.45Hurt
Use TDV³ instead of FOCNetDenoising · Urban100; σ=15, 25, 50PSNR: 33.15 vs 32.87; 30.64 vs 30.38; 27.40 vs 27.04Hurt
Use TDV³ instead of OISR-LF-sSingle-image super-resolution · BSDS100, ×2, ×3, and ×4PSNR: 32.18 vs 32.20; 28.98 vs 29.11; 27.50 vs 27.60No clear effect
Use TDV³ instead of OISR-LF-sSingle-image super-resolution · Set14, ×2, ×3, and ×4PSNR: 33.35 vs 33.62; 29.96 vs 30.35; 28.41 vs 28.63Hurt
Increase the number of TDV macro-blocks (ablation)TDV denoising · BSDS68, σ=25PSNRImproved
Increase discretization depth beyond S=10 (ablation)TDV denoising · BSDS68, σ=25PSNRNo clear effect
Apply a denoising-trained TDV regularizer to CT reconstruction without retrainingTDV³₂₅ · MAYO CT, 4-fold and 8-fold angular undersamplingreconstruction qualityImproved
Apply a denoising-trained TDV regularizer to MRI reconstruction without retrainingTDV³₂₅ · MRI, 4-fold and 6-fold Cartesian undersamplingreconstruction qualityImproved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy