Skip to content

A short script for doing inference with vgg + netvlad. The smaller the distance between the vlad vectors the more likely the images are of the same place. Useful for localisation and reverse image search.

Notifications You must be signed in to change notification settings

JoshuaChick/NetVLAD-pytorch-pretrained

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Run

git clone https://github.com/JoshuaChick/NetVLAD-pytorch-pretrained
cd NetVLAD-pytorch-pretrained
pip3 install -r requirements.txt
python script.py

Notes:

  • Put your image paths in img_tensor1 = preprocess_image('<image path 1>') and img_tensor2 = preprocess_image('<image path 2>') in script.py to run on your own images.
  • script.py will give you the distance between the NetVLAD vectors for two images (between 0 and 2, as vectors each have magnitude 1).

About

A short script for doing inference with vgg + netvlad. The smaller the distance between the vlad vectors the more likely the images are of the same place. Useful for localisation and reverse image search.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages