{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/guidedmix-net-learning-to-improve-pseudo","title":"GuidedMix-Net: Learning to Improve Pseudo Masks Using Labeled Images as Reference","arxiv_id":"2106.15064","date":"2021-06-29","proceeding":null,"authors":["Peng Tu","Yawen Huang","Rongrong Ji","Feng Zheng","Ling Shao"],"abstract":"Semi-supervised learning is a challenging problem which aims to construct a model by learning from a limited number of labeled examples. Numerous methods have been proposed to tackle this problem, with most focusing on utilizing the predictions of unlabeled instances consistency alone to regularize networks. However, treating labeled and unlabeled data separately often leads to the discarding of mass prior knowledge learned from the labeled examples, and failure to mine the feature interaction between the labeled and unlabeled image pairs. In this paper, we propose a novel method for semi-supervised semantic segmentation named GuidedMix-Net, by leveraging labeled information to guide the learning of unlabeled instances. Specifically, we first introduce a feature alignment objective between labeled and unlabeled data to capture potentially similar image pairs and then generate mixed inputs from them. The proposed mutual information transfer (MITrans), based on the cluster assumption, is shown to be a powerful knowledge module for further progressive refining features of unlabeled data in the mixed data space. To take advantage of the labeled examples and guide unlabeled data learning, we further propose a mask generation module to generate high-quality pseudo masks for the unlabeled data. Along with supervised learning for labeled data, the prediction of unlabeled data is jointly learned with the generated pseudo masks from the mixed data. Extensive experiments on PASCAL VOC 2012, PASCAL-Context and Cityscapes demonstrate the effectiveness of our GuidedMix-Net, which achieves competitive segmentation accuracy and significantly improves the mIoU by +7$\\%$ compared to previous state-of-the-art approaches.","url_abs":"https://arxiv.org/abs/2106.15064v2","url_pdf":"https://arxiv.org/pdf/2106.15064v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"guidedmix-net-learning-to-improve-pseudo","repo_url":"https://github.com/yh-pengtu/GuidedMix-Net","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"semi-supervised-semantic-segmentation","task_name":"Semi-Supervised Semantic Segmentation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-3","task":"Semi-Supervised Semantic Segmentation","dataset":"Cityscapes 100 samples labeled","model":"GuidedMix-Net(DeepLab v2 with ResNet101, ImageNet pretrained)","rank_in_archive_order":9,"of":13,"metrics":{"Validation mIoU":"56.9%"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-2","task":"Semi-Supervised Semantic Segmentation","dataset":"Cityscapes 12.5% labeled","model":"GuidedMix-Net(DeepLab v2 with ResNet101, ImageNet pretrained)","rank_in_archive_order":25,"of":33,"metrics":{"Validation mIoU":"65.8%"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-1","task":"Semi-Supervised Semantic Segmentation","dataset":"Cityscapes 25% labeled","model":"GuidedMix-Net(DeepLab v2 with ResNet101, ImageNet pretrained)","rank_in_archive_order":24,"of":30,"metrics":{"Validation mIoU":"67.5%"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-8","task":"Semi-Supervised Semantic Segmentation","dataset":"Cityscapes 50% labeled","model":"GuidedMix-Net(DeepLab v2 with ResNet101, ImageNet pretrained)","rank_in_archive_order":20,"of":23,"metrics":{"Validation mIoU":"69.8%"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-11","task":"Semi-Supervised Semantic Segmentation","dataset":"PASCAL Context 12.5% labeled","model":"GuidedMix-Net(DeepLab v2 with ResNet101, ImageNet pretrained)","rank_in_archive_order":1,"of":2,"metrics":{"Validation mIoU":"40.3%"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-12","task":"Semi-Supervised Semantic Segmentation","dataset":"PASCAL Context 25% labeled","model":"GuidedMix-Net(DeepLab v2 with ResNet101, ImageNet pretrained)","rank_in_archive_order":1,"of":2,"metrics":{"Validation mIoU":"41.7%"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-17","task":"Semi-Supervised Semantic Segmentation","dataset":"PASCAL VOC 2012 1000 labels","model":"GuidedMix-Net(DeepLab v2 with ResNet50, ImageNet pretrained)","rank_in_archive_order":1,"of":1,"metrics":{"Validation mIoU":"68.1%"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-10","task":"Semi-Supervised Semantic Segmentation","dataset":"PASCAL VOC 2012 1464 labels","model":"GuidedMix-Net(DeepLab v2 with ResNet50, ImageNet pretrained)","rank_in_archive_order":16,"of":17,"metrics":{"Validation mIoU":"73.7"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-9","task":"Semi-Supervised Semantic Segmentation","dataset":"PASCAL VOC 2012 25% labeled","model":"GuidedMix-Net(DeepLab v2 with ResNet101, input-size: 512x512 with multi-scale and flip, ImageNet pretrained)","rank_in_archive_order":15,"of":27,"metrics":{"Validation mIoU":"77.8%"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-9","task":"Semi-Supervised Semantic Segmentation","dataset":"PASCAL VOC 2012 25% labeled","model":"GuidedMix-Net(DeepLab v2 with ResNet101, ImageNet pretrained)","rank_in_archive_order":22,"of":27,"metrics":{"Validation mIoU":"75.5%"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-15","task":"Semi-Supervised Semantic Segmentation","dataset":"PASCAL VOC 2012 50%","model":"GuidedMix-Net(DeepLab v2 with ResNet101, input-size: 512x512 with multi-scale and flip, ImageNet pretrained)","rank_in_archive_order":7,"of":14,"metrics":{"Validation mIoU":"78.2%"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-15","task":"Semi-Supervised Semantic Segmentation","dataset":"PASCAL VOC 2012 50%","model":"GuidedMix-Net(DeepLab v2 with ResNet101, ImageNet pretrained)","rank_in_archive_order":11,"of":14,"metrics":{"Validation mIoU":"76.5%"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-16","task":"Semi-Supervised Semantic Segmentation","dataset":"PASCAL VOC 2012 500 labels","model":"GuidedMix-Net(DeepLab v2 with ResNet50, ImageNet pretrained)","rank_in_archive_order":1,"of":1,"metrics":{"Validation mIoU":"65.4%"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-4","task":"Semi-Supervised Semantic Segmentation","dataset":"Pascal VOC 2012 12.5% labeled","model":"GuidedMix-Net","rank_in_archive_order":16,"of":38,"metrics":{"Validation mIoU":"76.4%"},"uses_additional_data":false},{"leaderboard":"/sota/semi-supervised-semantic-segmentation-on-4","task":"Semi-Supervised Semantic Segmentation","dataset":"Pascal VOC 2012 12.5% labeled","model":"GuidedMix-Net","rank_in_archive_order":24,"of":38,"metrics":{"Validation mIoU":"73.4%"},"uses_additional_data":true}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2106.15064","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}