000101500 001__ 101500
000101500 005__ 20210507085643.0
000101500 0247_ $$2doi$$a10.1109/IROS.2018.8594185
000101500 0248_ $$2sideral$$a107640
000101500 037__ $$aART-2018-107640
000101500 041__ $$aeng
000101500 100__ $$0(orcid)0000-0003-4638-4655$$aAlonso Ruiz, Iñigo$$uUniversidad de Zaragoza
000101500 245__ $$aSemantic Segmentation from Sparse Labeling Using Multi-Level Superpixels
000101500 260__ $$c2018
000101500 5060_ $$aAccess copy available to the general public$$fUnrestricted
000101500 5203_ $$aSemantic segmentation is a challenging problemthat can benefit numerous robotics applications, since it pro-vides information about the content at every image pixel.Solutions to this problem have recently witnessed a boost onperformance and results thanks to deep learning approaches.Unfortunately, common deep learning models for semanticsegmentation present several challenges which hinder real lifeapplicability in many domains. A significant challenge is theneed of pixel level labeling on large amounts of trainingimages to be able to train those models, which implies avery high cost. This work proposes and validates a simplebut effective approach to train dense semantic segmentationmodels from sparsely labeled data. Labeling only a few pixelsper image reduces the human interaction required. We findmany available datasets, e.g., environment monitoring data, thatprovide this kind of sparse labeling. Our approach is basedon augmenting the sparse annotation to a dense one with theproposed adaptive superpixel segmentation propagation. Weshow that this label augmentation enables effective learning ofstate-of-the-art segmentation models, getting similar results tothose models trained with dense ground-truth.
000101500 536__ $$9info:eu-repo/grantAgreement/ES/DGA/T45-17R$$9info:eu-repo/grantAgreement/ES/MINECO-FEDER/DPI2015-69376-R$$9info:eu-repo/grantAgreement/ES/UZ/UZ2017-TEC-06
000101500 540__ $$9info:eu-repo/semantics/openAccess$$aby$$uhttp://creativecommons.org/licenses/by/3.0/es/
000101500 655_4 $$ainfo:eu-repo/semantics/article$$vinfo:eu-repo/semantics/acceptedVersion
000101500 700__ $$0(orcid)0000-0002-7580-9037$$aMurillo Arnal, Ana Cristina$$uUniversidad de Zaragoza
000101500 7102_ $$15007$$2520$$aUniversidad de Zaragoza$$bDpto. Informát.Ingenie.Sistms.$$cÁrea Ingen.Sistemas y Automát.
000101500 773__ $$g2018, 18401073 (2018), 5785-5792$$pProc. IEEE/RSJ Int. Conf. Intell. Rob. Syst.$$tProceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems$$x2153-0858
000101500 8564_ $$s3286918$$uhttps://zaguan.unizar.es/record/101500/files/texto_completo.pdf$$yPostprint
000101500 8564_ $$s3109547$$uhttps://zaguan.unizar.es/record/101500/files/texto_completo.jpg?subformat=icon$$xicon$$yPostprint
000101500 909CO $$ooai:zaguan.unizar.es:101500$$particulos$$pdriver
000101500 951__ $$a2021-05-07-08:04:19
000101500 980__ $$aARTICLE