AI Paper Atlas

Research map / Label-efficient and robust learning / Semi-supervised learning

Semi-Supervised Skin Detection by Network With Mutual Guidance

Yi He, Jiayuan Shi, Chuan Wang, Haibin Huang, Jiaming Liu, Guanbin Li, Risheng Liu, Jue Wang

ICCV 2019 · Label-efficient and robust learning › Semi-supervised learning · approach: GAN-based semi-supervised learning · 28 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 mutual guidance between skin and body decoders (UNet) (ablation)dual-task network · Own skin validation setIoU (%): 76.56 -> 80.11Improved
add mutual guidance (DeepLab-v3-MobileNet backbone) (ablation)dual-task network · Own skin validation setIoU (%): 75.15 -> 79.02Improved
add CRF loss to cross-entropy (ablation)semi-supervised loss · Own skin validation setIoU (%): 79.28 -> 79.48Improved
add WCE loss to cross-entropy (ablation)semi-supervised loss · Own skin validation setIoU (%): 79.28 -> 79.52Improved
add both CRF and WCE losses to cross-entropy (ablation)semi-supervised loss · Own skin validation setIoU (%): 79.28 -> 81.18Improved
reduce skin training data from 5k to 1k, keeping 5k bodyOurs · Own skin validation setIoU (%): 81.18 -> 75.29Hurt
use informative stage-1 guidance (body box, GMM skin) instead of zeros (ablation)dual-task network (UNet) · Own skin validation setIoU (%): 80.11 -> 80.74Improved
add gradient stopping from stage-2 guidance to stage-1 decoders (ablation)dual-task network · qualitative skin examplesmask qualityImproved
finetune stage 2 after stage-1 training rather than training from scratch (ablation)dual-task network · Own skin validation setIoUImproved
use shared encoder with extra body data, even without mutual guidance (ablation)dual-task network · Own skin validation setIoU (%): 76.74 vs 75.59Improved
use smaller DeepLab-v3-MobileNet backbone instead of UNet (about 60% of parameters) (ablation)dual-task network · Own skin validation setIoU (%): 80.11 -> 79.02Mixed

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