Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
IMAGE COMPRESSION METHOD AND APPARATUS, ELECTRONIC DEVICE, AND STORAGE MEDIUM
Document Type and Number:
WIPO Patent Application WO/2024/021387
Kind Code:
A1
Abstract:
The present disclosure relates to the technical field of image processing, in particular to an image compression method and apparatus, an electronic device, and a storage medium. The method provided by an embodiment of the present disclosure comprises: first segmenting an acquired original image to obtain a plurality of region images, and respectively determining the number of effective channels required by each region image during an image compression process, and then inputting the original image and the number of channels respectively corresponding to each region image of the original image into a trained spatially-adaptive compression auto-encoder, and obtaining a target latent representation and a target reconstructed image of the original image after neural network transformation. In the embodiment of the present disclosure, image compression is respectively carried out on each region image of the original image according to the number of the respective adaptive feature channels by means of the spatially-adaptive compression auto-encoder, so that computational efficiency in the image compression process may be significantly improved while ensuring that a decompressed target reconstructed image meets a certain quality, which can consider joint optimization of both distortion degree and encoding overhead.

Inventors:
GAO WEI (CN)
TAO LVFANG (CN)
LI GE (CN)
Application Number:
PCT/CN2022/133355
Publication Date:
February 01, 2024
Filing Date:
November 22, 2022
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
UNIV PEKING SHENZHEN GRADUATE SCHOOL (CN)
International Classes:
H04N19/189; G06N3/04; H04N19/119
Foreign References:
CN115278257A2022-11-01
CN112488297A2021-03-12
CN113919476A2022-01-11
US20200111238A12020-04-09
Attorney, Agent or Firm:
CHOFN INTELLECTUAL PROPERTY (CN)
Download PDF: