Pregled bibliografske jedinice broj: 958334
Convolutional Models for Segmentation and Localization
Convolutional Models for Segmentation and Localization // Engineering Power, 13 (2018), 2; 8-12 (podatak o recenziji nije dostupan, pregledni rad, stručni)
CROSBI ID: 958334 Za ispravke kontaktirajte CROSBI podršku putem web obrasca
Naslov
Convolutional Models for Segmentation and Localization
Autori
Krešo, Ivan ; Bevandić, Petra ; Oršić, Marin ; Šegvić, Siniša
Izvornik
Engineering Power (1331-7210) 13
(2018), 2;
8-12
Vrsta, podvrsta i kategorija rada
Radovi u časopisima, pregledni rad, stručni
Ključne riječi
Convolutional models
Sažetak
The revival of deep models has profoundly improved the accuracy of image classification models and provided a large improvement potential in related computer vision tasks. Recently, much attention has been directed to wards dense prediction models which produce distinct output in each image pixel. This paper addresses two particular instances of dense prediction: object localization and semantic segmentation. We briefly review the underlying operation principles, present some of our experimental results and discuss ways to analyze the success of learning and the utility of the resulting models.
Izvorni jezik
Engleski
Znanstvena područja
Računarstvo
POVEZANOST RADA
Ustanove:
Fakultet elektrotehnike i računarstva, Zagreb
