The absolute position information of the image content has always been a key issue discussed by people.  In the image convolution, people have been studying how to extract the absolute position information of the image, and there has been relatively great progress.  With the rapid development of Transformer in recent years, people began to compare two different feature extractors.  This article discusses the advantages and disadvantages of the two feature extractors starting from the absolute position information between the image features. And introduced the latest developments in the extraction of image position information for these two different networks in recent years