AOSSIE-Org / PictoPy

An Image sorter that sorts photos based on face encodings in it.
https://aossie-org.github.io/PictoPy/
20 stars 36 forks source link

Creating a test to validate the feature of captioning images retrieved from the database #61

Open kamisama-coder opened 5 days ago

kamisama-coder commented 5 days ago

made using the transformers class BlipForConditionalGeneration and the model name is salesforce/blip-image-captioning-base. the future scope involves : 1.Enhanced Caption Clustering in a Database Use unsupervised clustering algorithms, like DBSCAN or k-means, to organize images in the database based on their captions. Enable dynamic clustering updates when new images are added to the database.

  1. Query-Based Image Search Implement semantic search to allow users to find images using natural language queries. For example, a query like "A dog playing in a garden" should retrieve images with similar content.
Pranav0-0Aggarwal commented 5 days ago

great idea! Will add this in the project milestones for the upcoming year.

kamisama-coder commented 4 days ago

@Pranav0-0Aggarwal can you assign this issue to me

Pranav0-0Aggarwal commented 4 days ago

assigned

kamisama-coder commented 4 days ago

@Pranav0-0Aggarwal please review the PR

kamisama-coder commented 2 days ago

@Pranav0-0Aggarwal tell me changes to be made in PR.