15-09-2022 дата публикации
Номер: US20220292805A1
Принадлежит:
An image processing method and apparatus, and a device and a computer-readable storage medium. The method comprises: acquiring an image data set, wherein the image data set includes an image and accompanying text related to an unknown category in the image (S); and generating a probability and/or distribution of the unknown category by means of an unknown category acquisition model (S); wherein the probability and/or distribution of the unknown category comprises the probability of each pixel in the image being from the unknown category, the probability of the unknown category being present in the image, and a partitioning probability after the image is subdivided into a plurality of areas. By means of the method, a large amount of human labor costs and time can be saved on. 1. An image processing method , comprising:obtaining an image dataset, the image dataset comprising an image and an accompanying text related to an unseen class in the image; andgenerating, based on the image dataset, a probability and/or distribution of the unseen class using an unseen class obtaining model, the probability and/or distribution of the unseen class comprising a probability that each pixel in the image is from the unseen class, a probability that the unseen class is present in the image, and a regional probability after the image is subdivided into a plurality of regions.2. The image processing method of claim 1 , wherein claim 1 ,the unseen class obtaining model comprises a local branch, a semi-global branch, and a global branch,wherein the local branch is configured to generate the probability that each pixel in the image is from the unseen class based on annotation information of a seen class, the global branch is configured to generate the probability that the unseen class is present in the image based on the accompanying text, the semi-global branch is configured to generate the regional probability after the image is subdivided into the plurality of regions based on the ...
Подробнее