# Accuracy Improvement of Semi-Supervised Segmentation Using Supervised ClassMix and Sup-Unsup Feature Discriminator Source: [Accuracy Improvement of Semi-Supervised Segmentation Using Supervised ClassMix and Sup-Unsup Feature Discriminator]() Collection index: [llms.txt]() Data updated: 2026-04-10T05:00:13.311Z Authors: Takahiro Mano, Reiji Saito, Kazuhiro Hotta Published: 2026-04-08T14:18:00.000Z ## Abstract In semantic segmentation, the creation of pixel-level labels for training data incurs significant costs. To address this problem, semi-supervised learning, which utilizes a small number of labeled images alongside unlabeled images to enhance the performance, has gained attention. A conventional semi-supervised learning method, ClassMix, pastes class labels predicted from unlabeled images onto other images. However, since ClassMix performs operations using pseudo-labels obtained from unlabeled images, there is a risk of handling inaccurate labels. Additionally, there is a gap in data quality between labeled and unlabeled images, which can impact the feature maps. This study addresses these two issues. First, we propose a method where class labels from labeled images, along with the corresponding image regions, are pasted onto unlabeled images and their pseudo-labeled images. Second, we introduce a method that trains the model to make predictions on unlabeled images more similar to those on labeled images. Experiments on the Chase and COVID-19 datasets demonstrated an average improvement of 2.07% in mIoU compared to conventional semi-supervised learning methods. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Semantic Segmentation on Chase 1/8 labeled ratio](): Semantic segmentation; Chase 1/8 labeled ratio; Background IoU: 96.58. - [Semantic Segmentation on Chase 1/4 labeled ratio](): Semantic segmentation; Chase 1/4 labeled ratio; Background IoU: 96.63. - [Semantic Segmentation on COVID-19 1/8 labeled ratio](): Semantic segmentation; COVID-19 1/8 labeled ratio; IoU (Background): 96.72. - [Semantic Segmentation on COVID-19 1/4 labeled ratio](): Semantic segmentation; COVID-19 1/4 labeled ratio; Background IoU: 97.08. ## Original sources and implementation [Original paper]() [PDF]()