Could you take (embeddings of) frames from your query video and measure the pair...

		jointpdf on July 30, 2020 \| parent \| context \| favorite \| on: ScaNN: Efficient Vector Similarity Search Could you take (embeddings of) frames from your query video and measure the pairwise similarity to frames from the reference videos, then rank by some summary function of that list of similarity scores? What are the bottlenecks besides performance?