A method, apparatus and computer-readable storage medium for generating a picture title

Abstract

The embodiments of the present invention disclose a method, a device and a computer-readable storage medium for generating a picture title. The pictures and their corresponding document information are grabbed from a database; an encoding and decoding model in a neural network is used to generate a corresponding initial Title; according to a preset matching rule, calculate the matching degree between the initial title and the sentences in the corresponding document information to determine the title of the picture. Considering that the document information corresponding to the picture can more accurately reflect the specific characteristics of the picture, the picture is named in combination with the document information, so that the amount of title information obtained is more abundant, the description of the picture is more accurate, and the existing technology is effectively avoided. The title of the picture in the middle is too broad, and the repetition rate of the title of the same type of picture is high.

Description

technical field [0001] The present invention relates to the technical field of picture processing, and in particular, to a method, an apparatus and a computer-readable storage medium for generating a picture title. Background technique [0002] Image caption generation is a comprehensive problem that combines computer vision and natural language processing, that is, generating a caption for a picture to describe the information contained in the image. It is a research hotspot in recent years, and substantial progress has been made. At the same time, this is a very challenging task, we not only need to determine which entities are included in the image, but also express the relationship between them and generate reasonably fluent sentences. [0003] At present, the research methods of image caption generation can be roughly divided into the following two categories: one is the method combining visual detector and language modeling. The method first uses multi-instance learni...

AI Technical Summary

Problems solved by technology

[0004] However, the methods of generating picture titles in the prior art are all based on the characteristics of the picture itself, and often contain the following defects: the generated picture titles are too broad to provide specific information, such as: "a tennis player is playing tennis", " A building", but we have no way of knowing who the athlete is and which building it is

Moreover, the discrimination of the generated titles is poor, that is, the probability of generating the same title for similar pictures is high. For example, whether it is Federer or Nadal, they can only generate "a tennis player is playing tennis".

Images