AI Paper Atlas

Research map / Image generation / Image restoration

Tackling Ill-Posedness of Reversible Image Conversion With Well-Posed Invertible Network

Yuanfei Huang, Hua Huang

TPAMI 2026 · Image generation › Image restoration · approach: Inverse imaging with learned priors · 0 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
add L_shift to the WIC layer (ablation)WIN-Naive · DIV2K single hiding, cover/stegoPSNR: 37.09 -> 48.62Improved
add WIC layer (without L_shift) (ablation)INN baseline · DIV2K single hiding, cover/stegoPSNR: 37.34 -> 37.09Hurt
add WIC layer (without L_shift) (ablation)INN baseline · DIV2K single hiding, secret/recoverPSNR: 42.65 -> 44.22Improved
add L_Det on top of WIC and L_shift (ablation)WIN-Naive · DIV2K single hiding, cover/stegoPSNR: 48.62 -> 48.76Improved
add long-term memory (ablation)WIN · DIV2K single hiding, cover/stegoPSNR: 48.76 -> 50.05Improved
add long-term memory (ablation)WIN · DIV2K single hiding, secret/recoverPSNR: 49.13 -> 49.46Improved
lower lambda1 from 2 to 0.5 (ablation)WIN · DIV2K single hiding, cover/stegoPSNR: 48.76 -> 44.72Hurt
raise lambda2 from 1 to 4 (ablation)WIN · DIV2K single hiding, cover/stegoPSNR: 48.76 -> 45.21Hurt
raise lambda3 from 0.1 to 10 (ablation)WIN · DIV2K single hiding, cover/stegoPSNR: 48.76 -> 48.62No clear effect
raise lambda4 from 1 to 4 (ablation)WIN · DIV2K single hiding, cover/stegoPSNR: 48.76 -> 46.23Hurt
use WIN instead of StegFormer-B (30x fewer parameters)4-image hiding · DIV2K stegoPSNR: 40.18 -> 42.52Improved
use WIN instead of IRN for x2 rescalingimage rescaling · DIV2KPSNR: 40.87 -> 41.97Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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