US 12,633,104 B2 · Image feature transmission method, device and system

Abstract

The invention relates to an image feature transmission method, device and system, and relates to the technical field of communication. The transmission method includes the following steps: extracting a feature matrix of a to-be-processed image for each channel by using a machine learning model; determining one or more incidence matrix pairs according to a comparison result of the correlation degree between the feature matrixes and a first threshold value; according to the information amount, determining a representation matrix and a represented matrix in the two feature matrixes of each incidence matrix pair; determining a corresponding relation between each representation matrix and each represented matrix; and carrying out quantization processing and coding processing on each representation matrix, the corresponding relation and the maximum characteristic value and the minimum characteristic value in each represented matrix, and then transmitting to a decoding end.

Claims19

Dependent claims

A transmission method for image features, comprising:

extracting a feature matrix for a channel of an image to be processed using a machine learning model;
determining one or more pairs of correlated matrices based on a result of comparing a degree of correlation between feature matrices with a first threshold;
determining a representative matrix and a represented matrix from two feature matrices of each of the one or more pairs of correlated matrices based on an amount of information;
determining correspondence between the representative matrix and the represented matrix;
and
transmitting the representative matrix, the correspondence, and a maximum feature value and a minimum feature value of the represented matrix quantized and encoded to a decoder.

3 more elements

A transmission apparatus for image features, comprising:

a memory;
and
a processor coupled to the memory, the processor configured to, based on instructions stored in the memory, implement a transmission method for image features, comprising:
extracting a feature matrix for a channel of an image to be processed using a machine learning model;
determining one or more pairs of correlated matrices based on a result of comparing a degree of correlation between feature matrices with a first threshold;
determining a representative matrix and a represented matrix from two feature matrices of each of the one or more pairs of correlated matrices based on an amount of information;
determining correspondence between the representative matrix and the represented matrix;
and
transmitting the representative matrix, the correspondence, and a maximum feature value and a minimum feature value of the represented matrix quantized and encoded to a decoder.

6 more elements

Dependencies are inferred from the claim text.

Image feature transmission method, device and system

Life of this patent

Image feature transmission method, device and system

Life of this patent

Abstract

Claims19