Backbone network generation method and device, equipment and storage medium

Abstract

The invention provides a backbone network generation method and device, equipment and a storage medium, and relates to the technical field of artificial intelligence, in particular to a deep learning and computer vision technology. The method comprises the following steps: acquiring a training image set, a reasoning image set and an initial backbone network set; for each initial backbone network in the initial backbone network set, training and reasoning the initial backbone network by using the training image set and the reasoning image set to obtain reasoning time consumption and reasoning precision of the trained backbone network in the reasoning process; determining a basic backbone network based on the reasoning time consumption and reasoning precision of each trained backbone network in the reasoning process; and obtaining a target backbone network based on the basic backbone network and a preset target network. According to the backbone network generation method disclosed by the invention, the reasoning speed and reasoning precision of the backbone network on the Intel central processing unit are improved.

Description

technical field [0001] The present disclosure relates to the technical field of artificial intelligence, specifically deep learning and computer vision technology, and in particular to a generation method, device, equipment and storage medium of a backbone network. Background technique [0002] In computer vision tasks based on deep learning, such as image classification, object detection, image semantic segmentation, and metric learning, backbone networks are inseparable. As a feature extractor, the importance of backbone networks is self-evident. Some existing lightweight backbone networks, such as SHuffleNetV2, MobileNetV3, etc., these backbone networks are still not ideal for reasoning on the Intel CPU (Intel Central Processing Unit) side. As a result, tasks such as target detection and image segmentation cannot be done on the Intel CPU side. real-time predictions. Contents of the invention [0003] The present disclosure provides a method, device, equipment and stora...

AI Technical Summary

Problems solved by technology

Some existing lightweight backbone networks, such as SHuffleNetV2, MobileNetV3, etc., these backbone networks are still not ideal for reasoning on the Intel CPU (Intel Central Processing Unit) side. As a result, tasks such as target detection and image segmentation cannot be done on the Intel CPU side. real-time forecast

Images