search

caoyue10 / cvpr17-dvsq

The implementation of CVPR-17 paper "Deep Visual-Semantic Quantization of Efficient Image Retrieval"
70 stars 27 forks source link

When I retrieval images, my input is label and an image. Is that right? Does it contradicts with "CBIR" concept? #8

Closed xiaozhi2015 closed 6 years ago

xiaozhi2015 commented 6 years ago

@caoyue10 Looking forward to replys.

caoyue10 commented 6 years ago

CBIR is defined as that given image query (only image w/o label), return the images from the image database which are the most similar to the query.

xiaozhi2015 commented 6 years ago

Thx a lot! ^_^

xiaozhi2015 commented 6 years ago

But in Wiki I saw the description as ----"Content-based" means that the search analyzes the contents of the image rather than the metadata such as keywords, tags, or descriptions associated with the image. The term "content" in this context might refer to colors, shapes, textures, or any other information that can be derived from the image itself.
@caoyue10

xiaozhi2015 commented 6 years ago

BTW, how can I do image retrieval without any label infos using your code？

bl0 commented 6 years ago

Actually, You don't need the label for retrieval, we only use the label for performance evaluation.

xiaozhi2015 commented 6 years ago

Much thx! @bl0 And can I train the model without any label infos?