AI Paper Atlas

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

Hierarchical Information Passing Based Noise-Tolerant Hybrid Learning for Semi-Supervised Human Parsing

Yunan Liu, Shanshan Zhang, Jian Yang, PongChi Yuen

AAAI 2021 · Label-efficient and robust learning › Semi-supervised learning · approach: Confidence-based pseudo-label selection · 14 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 hierarchical information passing with a 3-stage network (ablation)HRNet human parsing · Extended PASCAL-Person-PartmIoU: 69.45 -> 73.51Improved
Add bottom-up information passing alongside top-down passing (ablation)HIPN · Extended PASCAL-Person-PartmIoU: 73.15 -> 73.51Improved
Increase HIPN stages from 1 to 4 with top-down and bottom-up passing (ablation)HIPN · Extended PASCAL-Person-PartmIoU: 71.81 -> 73.66Improved
Add cross-level consistency to certainty-based noisy-label identification (ablation)Hybrid learning for human parsing · Extended PASCAL-Person-PartmIoU: 74.57 -> 74.91Improved
Use certainty-based noise identification instead of NLNL identification (ablation)Hybrid learning for human parsing · Extended PASCAL-Person-Partclean/noisy classification accuracy: 60.59 -> 63.81Improved
Replace positive learning with noise-tolerant hybrid learning (ablation)Semi-supervised human parsing with 1/8 labeled LIP training data · LIP valmIoU: 42.91 -> 43.54Improved
Replace positive learning with noise-tolerant hybrid learning (ablation)Semi-supervised human parsing with 1/2 labeled LIP training data · LIP valmIoU: 53.97 -> 54.69Improved
Add unlabeled training images and hybrid learning to HIPNHIPN · LIP valmIoU: 59.61 -> 60.81Improved
Add unlabeled test images and hybrid learning to HIPNHIPN · Extended PASCAL-Person-Part testmIoU: 73.51 -> 74.91Improved

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