AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

Greatness in Simplicity: Unified Self-Cycle Consistency for Parser-Free Virtual Try-On

Chenghu Du, junyin Wang, Shuqing Liu, Shengwu Xiong

NeurIPS 2023 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Pretext-task visual pre-training · 4 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 auxiliary deformer GA in NGD (ablation)USC-PFN NGD · VITON unpairedFID: 42.23 -> 25.44Improved
add data term Edat in MRF (ablation)USC-PFN NGD · VITON unpairedFID: 30.53 -> 25.44Improved
add MRF module (ablation)USC-PFN SIG · VITON unpairedFID: 14.28 -> 10.47Improved
add Lsr skin supervision loss (ablation)USC-PFN SIG · VITON unpairedFID: 15.44 -> 10.47Improved
USC-PFN parser-free self-cycle vs in-painting SDAFNUSC-PFN · VITON pairedSSIM: 0.88 -> 0.91Improved
USC-PFN vs knowledge distillation PF-AFNUSC-PFN · VITON pairedSSIM: 0.89 -> 0.91Improved
USC-PFN vs cycle consistency DCTONUSC-PFN · VITON pairedSSIM: 0.83 -> 0.91Improved
USC-PFN vs cycle consistency DCTONUSC-PFN · VITON unpairedFID: 14.82 -> 10.47Improved
USC-PFN vs in-painting RT-VTONUSC-PFN · VITON unpairedFID: 11.66 -> 10.47Improved
USC-PFN vs knowledge distillation PF-AFNUSC-PFN · VITON unpairedFID: 10.09 -> 10.47Hurt
USC-PFN vs ACGPN inference speedUSC-PFN · VITONFPS: 10 -> 39Improved
USC-PFN vs DCTON FLOPsUSC-PFN · VITONFLOPs: 194G -> 46GImproved

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