Members-Only
Recent Talks & Demos are for members only
You must be an AI Tinkerers active member to view these talks and demos.
October 30, 2024
·
Cincinnati
ImageAI, YOLO, Llava Pipeline
See a demo of a toy that detects objects in camera feeds, crops them, and uses LLAVA for image processing with structured output.
Overview
I’m not gung-ho to do this, but I heard you’re looking for speakers. I built a little toy that utilizes A Python package called imageai to detect images in the camera (via a yolo3 model). It then performs some cropping on a detected image and sends it to llava (MS model) for some light image processing with structured output. I would be willing do demo it and walk through the source.
If that’s something of interest, I could talk about that. Probably would take no more than 20 minutes.
Links
Yolo-v3 detection crops objects for Llava multi-modal analysis and metadata extraction.
Tech stack