Finding products that are similar to a product selected from a plurality of products

A computer-implemented method comprising:

• receiving from a client machine a query for images similar to a desired image;
• computing a digest for the desired image, the digest including title and aspects of the desired image;
• responsive to information from the computed digest of the desired image, performing a similarity search of digests of other images for images similar to the desired image, the digests of other images including title and aspects of the other images, and computed previous to computing the digest for the desired image;
• and
• providing to the client machine at least one of the images that result from the similarity search.

One or more computer-readable hardware storage device having stored therein a set of instructions which, when executed by one or more processors of a computer, causes the computer to execute operations comprising:

• receiving from a client machine a query for images similar to a desired image;
• computing a digest for the desired image, the digest including title and aspects of the desired image;
• responsive to information from the computed digest of the desired image, performing a similarity search of digests of other images for images of similar to the desired image, the digests of other images including title and aspects of the other images, and computed previous to computing the digest for the desired image;
• and
• providing to the client machine at least one of the images that result from the similarity search.

A system comprising one or more hardware processors configured to:

• receive from a client machine a query for images similar to a desired image;
• compute a digest for the desired image, the digest including title and aspects of the desired image;
• responsive to information from the computed digest of the desired image, perform a similarity search of digests of other images for images of similar to the desired image, the digests of other images including title and aspects of the other images, and computed previous to computation of the digest for the desired image;
• and
• provide to the client machine at least one of the images that result from the similarity search.