Image processing method and apparatus
AssigneeHuawei Technologies Co., Ltd.
InventorsJin Song, Yin Zhao
Life of this patent
Estimating the term…
Abstract
An image processing method and apparatus to reduce storage resource consumption includes determining, by a server, a first geometric shape according to a first image unit in a first image from an encoding-side device, where a similarity between the first image unit and the first geometric shape falls within a preset range; and sending encoding information of the first image to a decoding-side device, where the encoding information of the first image includes indication information of the first geometric shape and indication information of a location of the first image unit in the first image.
Claims14
An image processing method implemented by a server, comprising:
- determining a first geometric shape in a preset geometric shape set according to each of a first image, the preset geometric shape set, and a preset similarity threshold;
- and
- determining a first image component in the first image, wherein the first image component corresponds to the first geometric shape, and wherein a similarity between the first image component and the first geometric shape is greater than or equal to the preset similarity threshold;
- generating first indication information that indicates information for the first geometric shape;
- sending the first indication information to a decoding-side device to enable the decoding-side device to reconstruct the first image;
- and
- sending the first indication information to an encoding-side device to enable the encoding-side device to:
- determine a first target geometric shape according to each of the first geometric shape, a second image, and the preset similarity threshold;
- determine a second image component in the second image;
- and
- generate second indication information that indicates a first target geometric shape, wherein the second image component corresponds to the first target geometric shape, and wherein a similarity between the second image component and the first target geometric shape is greater than or equal to the preset similarity threshold.
An image processing method implemented by a decoding-side device, comprising:
- receiving, from a server, first indication information that indicates a first geometric shape, wherein the first geometric shape is based on a preset geometric shape set that is based on each of a first image, the preset geometric shape set, and a preset similarity threshold, wherein the first image comprises a first image component, wherein the first image component corresponds to the first geometric shape, and wherein a similarity between the first image component and the first geometric shape is greater than or equal to the preset similarity threshold;
- determining the first geometric shape according to the first indication information;
- reconstructing the first image based on the first geometric shape;
- receiving, from the server, second indication information that indicates a second geometric shape, wherein the second geometric shape corresponds to a third image component, wherein the third image component is based on each of the first image, the preset geometric shape set, and a preset similarity threshold, wherein a similarity between the third image component and the second geometric shape is greater than or equal to the preset similarity threshold, and wherein a similarity between the third image component and each geometric shape in the preset geometric shape set is less than the preset similarity threshold;
- and
- determining the second geometric shape according to the second indication information, wherein reconstructing the first image based on the first geometric shape comprises reconstructing the first image according to the first geometric shape and the second geometric shape.
An image processing apparatus, comprising:
- a processor configured to:
- determine a first geometric shape in a preset geometric shape set according to each of a first image, the preset geometric shape set, and a preset similarity threshold;
- determine a first image component in the first image, wherein the first image component corresponds to the first geometric shape, and wherein a similarity between the first image component and the first geometric shape is greater than or equal to the preset similarity threshold;
- and
- generate first indication information that indicates the first geometric shape;
- a transmitter coupled to the processor and configured to:
- send the first indication information to a decoding-side device;
- and
- send the first indication information to an encoding-side device to enable an encoding-side device to:
- determine at least one first target geometric shape in the first geometric shape according to each of the first geometric shape, a second image, and the preset similarity threshold;
- determine a second image component in the second image;
- generate second indication information that indicates a first target geometric shape, wherein the second image component corresponds to the first target geometric shape, and wherein a similarity between the second image component and the first target geometric shape is greater than or equal to the preset similarity threshold.
An image processing apparatus, comprising:
- a receiver configured to receive, from a server, first indication information that indicates a first geometric shape, wherein the first geometric shape is determined from a preset geometric shape set according to each of a first image, the preset geometric shape set, and a preset similarity threshold, wherein the first image comprises a first image component, wherein the first image component corresponds to the first geometric shape, and wherein a similarity between the first image component and the first geometric shape is greater than or equal to the preset similarity threshold;
- a processor coupled to the receiver and configured to:
- determine the first geometric shape according to the first indication information;
- determine at least one first target geometric shape in the first geometric shape according to each of the first geometric shape, a second image, and the preset similarity threshold;
- determine a second image component in the second image, wherein the second image component corresponds to the first target geometric shape, and wherein a similarity between the second image component and the first target geometric shape is greater than or equal to the preset similarity threshold;
- and
- generate second indication information that indicates the first target geometric shape;
- and
- a transmitter coupled to the processor and configured to send the second indication information to a decoding-side device.
Dependencies are inferred from the claim text.