000108422 001__ 108422
000108422 005__ 20230526092852.0
000108422 0248_ $$2sideral$$a124898
000108422 0247_ $$2doi$$a10.1109/ICCV48922.2021.00811
000108422 037__ $$aART-2021-124898
000108422 041__ $$aeng
000108422 100__ $$0(orcid)0000-0003-4638-4655$$aAlonso, Iñigo$$uUniversidad de Zaragoza
000108422 245__ $$aSemi-Supervised Semantic Segmentation with Pixel-Level Contrastive Learning from a Class-wise Memory Bank
000108422 260__ $$c2021
000108422 5060_ $$aAccess copy available to the general public$$fUnrestricted
000108422 5203_ $$aThis work presents a novel approach for semi-supervised semantic segmentation. The key element of this approach is our contrastive learning module that enforces the segmentation network to yield similar pixel-level feature representations for same-class samples across the whole dataset. To achieve this, we maintain a memory bank continuously updated with relevant and high-quality feature vectors from labeled data. In an end-to-end training, the features from both labeled and unlabeled data are optimized to be similar to same-class samples from the memory bank. Our approach outperforms the current state-of-the-art for semi-supervised semantic segmentation and semi-supervised domain adaptation on well-known public benchmarks, with larger improvements on the most challenging scenarios, i.e., less available labeled data.
000108422 536__ $$9info:eu-repo/grantAgreement/ES/DGA/T45-17R$$9info:eu-repo/grantAgreement/ES/MINECO-FEDER/PGC2018-098817-A-I00
000108422 540__ $$9info:eu-repo/semantics/openAccess$$aby$$uhttp://creativecommons.org/licenses/by/3.0/es/
000108422 655_4 $$ainfo:eu-repo/semantics/article$$vinfo:eu-repo/semantics/acceptedVersion
000108422 700__ $$aSabater, Alberto$$uUniversidad de Zaragoza
000108422 700__ $$aFerstl, David
000108422 700__ $$0(orcid)0000-0003-1183-349X$$aMontesano, Luis$$uUniversidad de Zaragoza
000108422 700__ $$0(orcid)0000-0002-7580-9037$$aMurillo, Ana Cristina$$uUniversidad de Zaragoza
000108422 7102_ $$15007$$2520$$aUniversidad de Zaragoza$$bDpto. Informát.Ingenie.Sistms.$$cÁrea Ingen.Sistemas y Automát.
000108422 7102_ $$15007$$2570$$aUniversidad de Zaragoza$$bDpto. Informát.Ingenie.Sistms.$$cÁrea Lenguajes y Sistemas Inf.
000108422 773__ $$g2021 (2021), 8219-8228$$pProceedings (IEEE International Conference on Computer Vision)$$tProceedings (IEEE International Conference on Computer Vision)$$x1550-5499
000108422 85641 $$uhttps://arxiv.org/abs/2104.13415$$zTexto completo de la revista
000108422 8564_ $$s2314946$$uhttps://zaguan.unizar.es/record/108422/files/texto_completo.pdf$$yPostprint
000108422 8564_ $$s2492209$$uhttps://zaguan.unizar.es/record/108422/files/texto_completo.jpg?subformat=icon$$xicon$$yPostprint
000108422 909CO $$ooai:zaguan.unizar.es:108422$$particulos$$pdriver
000108422 951__ $$a2023-05-26-08:13:32
000108422 980__ $$aARTICLE