Autor:

Gao, Depeng ; Liu, Jiafeng ; Wu, Rui ; Cheng, Dansong ; Fan, Xiaopeng ; Tang, Xianglong

Współtwórca:

Kulczycki, Piotr - ed. ; Kacprzyk, Janusz - ed. ; Kóczy, László T. - ed. ; Mesiar, Radko - ed.

Tytuł:

Utilizing relevant RGB-D data to help recognize RGB images in the target domain

Podtytuł:

.

Tytuł publikacji grupowej:

AMCS, volume 29 (2019)

Temat i słowa kluczowe:

object recognition ; RGB-D images ; transfer learning ; privileged information

Abstract:

With the advent of 3D cameras, getting depth information along with RGB images has been facilitated, which is helpful in various computer vision tasks. However, there are two challenges in using these RGB-D images to help recognize RGB images captured by conventional cameras: one is that the depth images are missing at the testing stage, the other is that the training and test data are drawn from different distributions as they are captured using different equipment. To jointly address the two challenges, we propose an asymmetrical transfer learning framework, wherein three classifiers are trained using the RGB and depth images in the source domain and RGB images in the target domain with a structural risk minimization criterion and regularization theory.

Wydawca:

Zielona Góra: Uniwersytet Zielonogórski

Data wydania:

2019

Typ zasobu:

artykuł

DOI:

10.2478/amcs-2019-0045

Strony:

611-621

Źródło:

AMCS, volume 29, number 3 (2019) ; kliknij tutaj, żeby przejść

Jezyk:

eng

Licencja CC BY 4.0:

kliknij tutaj, żeby przejść

Prawa do dysponowania publikacją:

Biblioteka Uniwersytetu Zielonogórskiego