AI Paper Atlas

Research map / Image generation / Image restoration

Invertible Denoising Network: A Light Solution for Real Noise Removal

Yang Liu, Zhenyue Qin, Saeed Anwar, Pan Ji, Dongwoo Kim, Sabrina Caldwell, Tom Gedeon

CVPR 2021 · Image generation › Image restoration · approach: Self-supervised image denoising · 197 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
replace the sampled clean-detail latent with a prior sample while retaining the low-resolution clean representationInvDN · SIDD testPSNRImproved
use InvDN instead of DANetreal-noise image denoising · SIDD testPSNR: 39.25 -> 39.28Improved
use InvDN instead of MIRNetreal-noise image denoising · DNDPSNR: 39.88 -> 39.57Hurt
use Haar wavelet instead of squeeze for down-scaling (ablation)InvDN · SIDD validationPSNRImproved
use residual blocks instead of dense blocks for the coupling subnetworks (ablation)InvDN · SIDD denoisingaccuracyImproved
use residual blocks instead of dense blocks for the coupling subnetworks (ablation)InvDN · SIDD denoisingnumber of parameters: 4.3M -> 2.6MImproved
increase invertible blocks from 8 to 16 (ablation)InvDN · SIDD validationPSNR: 38.40 -> 38.61Improved
use 2 down-scale blocks instead of 1 or 3 (ablation)InvDN with 8 invertible blocks · SIDD validationPSNR: 38.00, 38.40, 37.69Mixed
use 2 down-scale blocks instead of 1 or 3 (ablation)InvDN with 16 invertible blocks · SIDD validationPSNR: 38.30, 38.61, 38.15Mixed
use L2 for the forward low-resolution objective and L1 for the backward reconstruction objective (ablation)InvDN · SIDD validationPSNR: 38.88Improved
add gradient or SSIM loss (ablation)InvDN · SIDD validationPSNR: 38.74, 38.80, 38.78No clear effect
average outputs from 16 sampled latent variables using Monte Carlo self-ensembleInvDN · SIDD validationimages with a performance boost: 83.35%Improved

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