Visual relation detection method and device based on scene graph high-order semantic structure

Abstract

The invention provides a visual relation detection method and device based on a high-order semantic structure of a scene graph, and the algorithm comprises the steps: predicting the types and positions of all objects in a picture, outputting a visual feature vector corresponding to each object, and carrying out the pairing operation of every two of all detected objects; based on a pairing result, extracting a combined visual feature vector, and encoding the position to obtain a position code; inputting the categories of all the objects into a hierarchical semantic clustering algorithm, and processing to obtain a high-level semantic feature vector corresponding to each object; performing semantic coding on the output of the hierarchical semantic clustering algorithm; generating a relation classifier weight; and combining the visual feature vector, the joint visual feature vector and the position code into a unified feature vector, carrying out point product operation on the unified feature vector by using the relation classifier weight, and finally obtaining a relation conditional probability between every two objects as a scene graph.

Description

technical field [0001] The invention relates to the field of image processing, in particular to a visual relationship detection method and device based on a high-order semantic structure of a scene graph. Background technique [0002] The main goal of the visual relationship detection task is to identify and localize the content of the visual triplet relationship (subject, relation, object) present in the image. Recognition refers to identifying the category attributes of the target object, and positioning is to return the bounding box of the target object. Understanding a visual scene is often more than just recognizing individual objects, and even a perfect object detector would struggle to perceive the slight difference between a person feeding a horse and a person standing next to it. Learning rich semantic relationships among these objects is what visual relationship detection is all about. The key to a deeper understanding of the visual scene is to build a structured...

AI Technical Summary

Problems solved by technology

[0007] In order to solve the technical problem that it is difficult to obtain the correct type and quantity of annotations and complete triplet annotations, the present invention proposes a visual relationship detection method and device based on the high-order semantic structure of the scene graph

Images