Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
DATA COMPRESSION METHOD AND RELATED DEVICE
Document Type and Number:
WIPO Patent Application WO/2023/174256
Kind Code:
A1
Abstract:
The present application relates to the field of artificial intelligence. Disclosed in the present application is a data compression method, comprising: acquiring first target data, the first target data comprising first sub-data and second sub-data; according to the first sub-data, obtaining a first probability distribution by means of a first decoder of a variational autoencoder, the first probability distribution being used as a conditional probability distribution of the second sub-data; according to the first probability distribution, compressing the second sub-data by means of an entropy encoder, so as to obtain a first bit stream; and compressing the first sub-data to the first bit stream, so as to obtain a second bit stream. Compared with a bits-back coding mechanism requiring additionally arranged initial bits in the prior art, the embodiments of the present application do not need additionally arranged initial bits, can realize compression of a single data point, and reduce the compression ratio of parallel compression.

Inventors:
ZHANG CHEN (CN)
RYDER TOM (CN)
KANG NING (CN)
ZHANG SHIFENG (CN)
Application Number:
PCT/CN2023/081315
Publication Date:
September 21, 2023
Filing Date:
March 14, 2023
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
HUAWEI TECH CO LTD (CN)
International Classes:
H04L69/04
Foreign References:
CN110290387A2019-09-27
CN113810058A2021-12-17
US20200104640A12020-04-02
CN113569243A2021-10-29
Other References:
ZHAI, ZHENGLI; LIANG, ZHENMING; ZHOU, WEI; SUN, XIA: "Research Overview of Variational Auto-Encoders Models", COMPUTER ENGINEERING AND APPLICATIONS, HUABEI JISUAN JISHU YANJIUSUO, CN, vol. 55, no. 3, 1 February 2019 (2019-02-01), CN , pages 1 - 9, XP009548958, ISSN: 1002-8331, DOI: 10.3778/j.issn.1002-8331.1810-0284
Attorney, Agent or Firm:
SHENPAT INTELLECTUAL PROPERTY AGENCY (CN)
Download PDF: