this post was submitted on 12 Nov 2023
1 points (100.0% liked)

Machine Learning

1 readers
1 users here now

Community Rules:

founded 11 months ago
MODERATORS
 

Hey there, Reddit!

I'm part of a team at an app company that showcases the latest offers from supermarkets and hypermarkets. Our current workflow involves manually sorting lots of flyers into 14 categories and individually cropping images from these flyers which a team of data entry operators are currently doing. It's as time-intensive as it sounds.

I'm curious about leveraging machine learning to streamline this. My research suggests that ResNet50 could be a strong contender for the image classification piece. Has anyone here applied it in a similar context, or could you recommend something more apt or better for this task?

Regarding image cropping, is there an ML approach that could automate this step, recognizing and cutting out the pertinent sections from a variety of flyer layouts?

I'm a software engineer and as such have limited knowledge of working with ml, however, my current plan is to develop it with the help of chatgpt. In the past, I had built a simple stock price prediction system using lstm. Any shared experiences, alternative suggestions, or resources would be immensely valuable.

Looking forward to your input!

you are viewing a single comment's thread
view the rest of the comments
[–] mcjoness@alien.top 1 points 10 months ago

Have you done any research into image cropping techniques? Read white papers, pubs?