search

whitphx / lear-gist-python

A python wrapper for Lear's GIST implementation working with numpy
MIT License
55 stars 14 forks source link

Visualise extracted gist features #26

Open kk2491 opened 5 years ago

kk2491 commented 5 years ago

Hi,

Thanks for this repository, it wa really helpful in my RnD work.

I was able to successfully install and extract the gist features. I would like to understand, is there any way I can visualise these extracted gist features on the image. This would be really useful as we can see which are the image segments are extracted and how they are going to effect the machine learning model which I am planning to implement on the extracted features.

Thank you, KK

whitphx commented 5 years ago

In my understanding,

which are the image segments are extracted

This is what nblocks parameter specifies. The image is segmented into grid of nblocks * nblocks cells and the features are extracted from those cells separately.

But,

how they are going to effect the machine learning model which I am planning to implement on the extracted features.

I don't have any idea on this...