AI Paper Atlas

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

Multi-Granularity Distillation Scheme towards Lightweight Semi-Supervised Semantic Segmentation

Jie Qin, Jie Wu, Ming Li, Xuefeng Xiao, Min Zheng, Xingang Wang

ECCV 2022 · Label-efficient and robust learning › Semi-supervised learning · approach: Consistency training for semi-supervised segmentation · 10 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 CPS with MGD distillationResNet-18 student · VOC2012 val (1/16 labeled)mIoU: 47.13 -> 66.86Improved
replace KD with MGD distillationResNet-18 student · VOC2012 val (1/16 labeled)mIoU: 56.37 -> 66.86Improved
replace CPS with MGDResNet-18 student · Cityscapes val (1/16 labeled)mIoU: 59.43 -> 70.91Improved
replace CPS with MGDMobileNetv2 student · Cityscapes val (1/16 labeled)mIoU: 53.34 -> 58.00Improved
add pixel-level consistency loss to supervised training (ablation)MGD ResNet-18 · VOC2012 val (1/8 labeled)mIoU: 55.65 -> 63.61Improved
add image-level semantic loss from T_D (ablation)MGD ResNet-18 · VOC2012 val (1/8 labeled)mIoU: 63.61 -> 66.01Improved
add region-level content-aware loss from T_W (ablation)MGD ResNet-18 · VOC2012 val (1/8 labeled)mIoU: 63.61 -> 66.02Improved
add both image-level loss on T_D and region-level loss on T_W (ablation)MGD ResNet-18 · VOC2012 val (1/8 labeled)mIoU: 63.61 -> 68.29Improved
apply image- and region-level losses to both teachers instead of one each (ablation)MGD ResNet-18 · VOC2012 val (1/8 labeled)mIoU: 68.29 -> 67.51Hurt
use complementary T_D plus T_W instead of two T_D teachers (ablation)MGD ResNet-18 · VOC2012 val (1/8 labeled)mIoU: 66.14 -> 68.29Improved
use two teachers instead of a single T_D (ablation)MGD ResNet-18 · VOC2012 val (1/8 labeled)mIoU: 64.85 -> 68.29Improved
add unlabeled data distillation losses (ablation)MGD ResNet-18 · VOC2012 val (1/8 labeled)mIoU: 58.24 -> 68.29Improved

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