AI Paper Atlas

Research map / Image generation / Editing and style transfer

Do Not Mask What You Do Not Need to Mask: a Parser-Free Virtual Try-On

Thibaut Issenhuth, Jérémie Mary, Clément Calauzènes

ECCV 2020 · Image generation › Editing and style transfer · approach: Virtual try-on · 120 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 agnostic person representation of CP-VTON with parser-free T-WUTON inputCP-VTON · Virtual try-on test setFID: 37.237 -> 16.843Improved
train with teacher-student parser-free pipeline (T-WUTON)CP-VTON on p*_t-wuton · Virtual try-on test setFID: 16.843 -> 9.877Improved
distill teacher into parser-free student (S-WUTON)T-WUTON · Virtual try-on test setFID: 9.877 -> 7.927Improved
distill into student removing parser and pose at inferenceT-WUTON · Runtime (V100)total runtime (ms): 181ms -> 13msImproved
add adversarial loss to student (ablation)S-WUTON · Virtual try-on test setFID: 12.620 -> 7.927Improved
add adversarial loss to student (ablation)S-WUTON · Virtual try-on test setIS: 2.912 -> 3.154Improved
add unpaired adversarial loss (ablation)T-WUTON · Virtual try-on test setFID: 13.020 -> 9.877Improved
add unpaired adversarial loss (ablation)T-WUTON · Virtual try-on test setIS: 2.729 -> 3.114Improved
add unpaired adversarial loss (ablation)T-WUTON · Virtual try-on test setLPIPS: 0.107 -> 0.101Improved
train geometric matcher end-to-end with synthesis loss (ablation)T-WUTON · Virtual try-on test setFID: 11.125 -> 9.877Improved
train geometric matcher end-to-end (ablation)T-WUTON · Virtual try-on test setLPIPS: 0.112 -> 0.101Improved
train geometric matcher end-to-end (ablation)T-WUTON · Virtual try-on test setIS: 3.102 -> 3.114No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 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