Large-scale Image Retrieval Based on Deep Semantics
With the development of digital technology,the number of images has increased exponentially.Image retrieval has become more and more important.Therefore,this paper aims to use the image semantic feature to complete image retrieval.Firstly,the MixNet is used to obtain the deep semantics and then the vector database Milvus is applied.The results on the caltech-101 dataset show that the method applied in this paper can represent the image information well and achieve high-performance image retrieval.
NETL
NSTL
万方数据
