MARC보기
LDR02176nmm uu200445 4500
001000000333380
00520240805172923
008181129s2018 |||||||||||||||||c||eng d
020 ▼a 9780438324800
035 ▼a (MiAaPQ)AAI10816767
035 ▼a (MiAaPQ)berkeley:17873
040 ▼a MiAaPQ ▼c MiAaPQ ▼d 248032
0820 ▼a 001.5
1001 ▼a Zhang, Richard.
24510 ▼a Image Synthesis for Self-supervised Visual Representation Learning.
260 ▼a [S.l.] : ▼b University of California, Berkeley., ▼c 2018
260 1 ▼a Ann Arbor : ▼b ProQuest Dissertations & Theses, ▼c 2018
300 ▼a 138 p.
500 ▼a Source: Dissertation Abstracts International, Volume: 80-01(E), Section: B.
500 ▼a Adviser: Alexei A. Efros.
5021 ▼a Thesis (Ph.D.)--University of California, Berkeley, 2018.
520 ▼a Deep networks are extremely adept at mapping a noisy, high-dimensional signal to a clean, low-dimensional target output (e.g., image classification). By solving this heavy compression task, the network also learns about natural image priors. How
520 ▼a Part I describes the use of deep networks for conditional image synthesis. The section begins by exploring the problem of image colorization, proposing both automatic and user-guided approaches. This section then proposes a system for general im
520 ▼a Part II explores the visual representations learned within deep networks. Colorization, as well as cross-channel prediction in general, is a simple but powerful pretext task for self-supervised learning. The representations from cross-channel pr
590 ▼a School code: 0028.
650 4 ▼a Artificial intelligence.
650 4 ▼a Electrical engineering.
650 4 ▼a Computer science.
690 ▼a 0800
690 ▼a 0544
690 ▼a 0984
71020 ▼a University of California, Berkeley. ▼b Electrical Engineering & Computer Sciences.
7730 ▼t Dissertation Abstracts International ▼g 80-01B(E).
773 ▼t Dissertation Abstract International
790 ▼a 0028
791 ▼a Ph.D.
792 ▼a 2018
793 ▼a English
85640 ▼u http://www.riss.kr/pdu/ddodLink.do?id=T14998287 ▼n KERIS
980 ▼a 201812 ▼f 2019
990 ▼a 관리자