Similar Image Retrieval using Autoencoders

jkl133·2021년 2월 26일
0

How can we utilize the autoencoder for practical purposes?

  • It turns out that encoded representations (embeddings) given by the encoder are magnificent objects for similarity retrieval.
  • With such a condensed encoding representation, simple vector similarity measures between embeddings (such as cosine similarity) will create much more human-interpretable similarities between images.

https://github.com/ankonzoid/artificio/tree/master/image_retrieval

https://pdfs.semanticscholar.org/b324/85915bdf26e0e0d54530c7bf96d1c35c6822.pdf?_ga=2.197163814.841600703.1611907930-1544509872.1611907930https://pdfs.semanticscholar.org/b324/85915bdf26e0e0d54530c7bf96d1c35c6822.pdf?_ga=2.197163814.841600703.1611907930-1544509872.1611907930

profile
재밌는게 재밌는거다

0개의 댓글