Title:
METHOD AND APPARATUS FOR COMPRESSION AND TRAINING OF NEURAL NETWORK
Document Type and Number:
WIPO Patent Application WO/2021/261950
Kind Code:
A1
Abstract:
A neural-network-based signal processing method and apparatus according to the present invention may: receive a bitstream including information about a neural network model, wherein the bitstream includes at least one neural network access unit; obtain information about the at least one neural network access unit from the bitstream; and reconstruct the neural network model on the basis of the information about the at least one neural network access unit.
More Like This:
Inventors:
AHN YONGJO (KR)
LEE JONGSEOK (KR)
LEE JONGSEOK (KR)
Application Number:
PCT/KR2021/008005
Publication Date:
December 30, 2021
Filing Date:
June 25, 2021
Export Citation:
Assignee:
INTELLECTUAL DISCOVERY CO LTD (KR)
International Classes:
G06T9/00; G06N3/04; G06N3/08; G06N5/04
Domestic Patent References:
WO2020061008A1 | 2020-03-26 |
Foreign References:
KR20200005402A | 2020-01-15 | |||
KR20200052200A | 2020-05-14 | |||
KR20190049552A | 2019-05-09 |
Other References:
ANONYMOUS: "Working Draft 4 of Compression of neural networks for multimedia content description and analysis ", ISO/IEC JTC 1/SC 29/WG 11N19225, ISO/IEC JTC 1/SC 29/WG 11, 15 May 2020 (2020-05-15), XP055882959
Attorney, Agent or Firm:
CHOE, Yun Seo (KR)
Download PDF: