[D] Online (realtime) image clustering

Hey, I faced an unusual task and I'm not sure how to implement it.

Let's say I have a DB with a lot of images (with possible duplicates). First, I calculate embeddings for each of them with some encoder (irrelevant) and then apply clustering algorithm on these embeddings. The most important part is that I need to assign the cluster ID to each image.

Now, the tricky part is: new images are coming in to the system and I want …

algorithm apply clustering clustering algorithm embeddings encoder hey image images machinelearning part realtime