Mostra el registre d'ítem simple
Refinement network for unsupervised on the scene foreground segmentation
dc.contributor.author | Pardàs Feliu, Montse |
dc.contributor.author | Canet Tarrés, Gemma |
dc.contributor.other | Universitat Politècnica de Catalunya. Departament de Teoria del Senyal i Comunicacions |
dc.contributor.other | Universitat Politècnica de Catalunya. Doctorat en Teoria del Senyal i Comunicacions |
dc.date.accessioned | 2020-11-17T14:34:07Z |
dc.date.available | 2020-11-17T14:34:07Z |
dc.date.issued | 2020 |
dc.identifier.citation | Pardàs, M.; Canet, G. Refinement network for unsupervised on the scene foreground segmentation. A: European Signal Processing Conference. "28th European Signal Processing Conference (EUSIPCO 2020): 24-28 August 2020: Amsterdam, the Netherlands". European Association for Signal Processing (EURASIP), 2020, p. 705-709. ISBN 978-9-0827-9705-3. |
dc.identifier.isbn | 978-9-0827-9705-3 |
dc.identifier.uri | http://hdl.handle.net/2117/332324 |
dc.description.abstract | Unsupervised learning represents one of the most interesting challenges in computer vision today. The task has an immense practical value with many applications in artificial intelligence and emerging technologies, as large quantities of unlabeled images and videos can be collected at low cost. In this paper, we address the unsupervised learning problem in the context of segmenting the main foreground objects in single images. We propose an unsupervised learning system, which has two pathways, the teacher and the student, respectively. The system is designed to learn over several generations of teachers and students. At every generation the teacher performs unsupervised object discovery in videos or collections of images and an automatic selection module picks up good frame segmentations and passes them to the student pathway for training. At every generation multiple students are trained, with different deep network architectures to ensure a better diversity. The students at one iteration help in training a better selection module, forming together a more powerful teacher pathway at the next iteration. In experiments, we show that the improvement in the selection power, the training of multiple students and the increase in unlabeled data significantly improve segmentation accuracy from one generation to the next. Our method achieves top results on three current datasets for object discovery in video, unsupervised image segmentation and saliency detection. At test time, the proposed system is fast, being one to two orders of magnitude faster than published unsupervised methods. We also test the strength of our unsupervised features within a well known transfer learning setup and achieve competitive performance, proving that our unsupervised approach can be reliably used in a variety of computer vision tasks. |
dc.description.sponsorship | During the development of this work the first author was a visitor at TOSHIBA Cambridge Research Lab. This work has been carried out with the support of this lab and project TEC2016-75976-R, by the Ministerio de Economia, Industria y Competitividad and the European Regional Development Fund. |
dc.format.extent | 5 p. |
dc.language.iso | eng |
dc.publisher | European Association for Signal Processing (EURASIP) |
dc.subject | Àrees temàtiques de la UPC::Enginyeria de la telecomunicació::Processament del senyal::Processament de la imatge i del senyal vídeo |
dc.subject.lcsh | Machine learning |
dc.subject.lcsh | Neural networks (Computer science) |
dc.subject.lcsh | Image processing |
dc.subject.other | Unsupervised learning |
dc.subject.other | Foreground object segmentation |
dc.subject.other | Object discovery in video |
dc.subject.other | Transfer learning |
dc.title | Refinement network for unsupervised on the scene foreground segmentation |
dc.type | Conference report |
dc.subject.lemac | Aprenentatge automàtic |
dc.subject.lemac | Xarxes neuronals (Informàtica) |
dc.subject.lemac | Imatges -- Processament |
dc.contributor.group | Universitat Politècnica de Catalunya. GPI - Grup de Processament d'Imatge i Vídeo |
dc.description.peerreviewed | Peer Reviewed |
dc.relation.publisherversion | https://www.eurasip.org/Proceedings/Eusipco/Eusipco2020/pdfs/0000705.pdf |
dc.rights.access | Open Access |
local.identifier.drac | 28860934 |
dc.description.version | Postprint (published version) |
dc.relation.projectid | info:eu-repo/grantAgreement/MINECO/1PE/TEC2016-75976-R |
local.citation.author | Pardàs, M.; Canet, G. |
local.citation.contributor | European Signal Processing Conference |
local.citation.publicationName | 28th European Signal Processing Conference (EUSIPCO 2020): 24-28 August 2020: Amsterdam, the Netherlands |
local.citation.startingPage | 705 |
local.citation.endingPage | 709 |