UMass-Rescue / UMass-Rescue/RescueBox

Plugin explore image caption models

Open
#73 0 comments 0 reactions 1 assignee View on GitHub

@jelchou is already working on this.

Since Oct 24, 2025.

good first issue hackathon
Dominant language
Python
Stars
6
Forks
15
Avg merge
14h 42m
Merged PRs (30d)
7

Description

Image Summary Plugin uses LLM mutimodal feature to describe images. however this needs GPU

can we use a pipeline approach to first scan images get a caption , for example find all images with a face or a white car and then this subset can feed into above Image summary if more details is desired.

so goal here is get a caption from an image using less GPU intensive approach.

Use CLIP for the initial, flexible, zero-shot searching and filtering of image databases. This helps investigators quickly narrow down a massive amount of data.

Use a lightweight version of LLaVA for conversational, higher-level analysis, such as generating descriptions or answering questions about specific images that were flagged by CLIP.

bliva

lavis
and
https://github.com/salesforce/BLIP

refer rescuebox wiki and hello-world-plugin

DETR

vision transformers

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.