AI Paper Atlas

Research map / Learning theory and algorithms / Matrix and tensor methods

Deeply Learned Spectral Total Variation Decomposition

Tamara G. Grossmann, Yury Korolev, Guy Gilboa, Carola Schoenlieb

NeurIPS 2020 · Learning theory and algorithms › Matrix and tensor methods · approach: Tensor network decompositions · 2 citations (OpenAlex)

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 changedWhereMeasuredEffect
Replace the repeated model-driven TV-flow optimization with a trained TVspecNETTV spectral decomposition · MS COCO test, 1024 × 1024 pixelscomputation time (seconds): 28.2811 -> 0.0020Improved
Use an FFDnet basis network instead of DnCNN (ablation)TVspecNET · MS COCO testSSIM: 0.9600 -> 0.9377Hurt
Use a U-Net basis network instead of DnCNN (ablation)TVspecNET · MS COCO testSSIM: 0.9600 -> 0.9233Hurt
Add the band-sum reconstruction loss to MSE (ablation)TVspecNET · MS COCO testSSIM: 0.9600 -> 0.9639No clear effect
Add the normalized gradient Huber loss to MSE (ablation)TVspecNET · MS COCO testSSIM: 0.9600 -> 0.9559No clear effect
Add band-sum reconstruction and gradient Huber losses to MSE (ablation)TVspecNET · MS COCO testSSIM: 0.9600 -> 0.9619No clear effect
Train additionally on isolated disk eigenfunctions (ablation)TVspecNET · Disk imagesspectral band recoveryNo clear effect
Train additionally on disk images (ablation)TVspecNET · MS COCO testSSIM: 0.9600 -> 0.9533Hurt

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