Inspiration
I wanted to create a program that is capable of identifying an object with visual information, just like human eye.
What it does
This program can identify a body part by an image or as a moving object with webcam.
How we built it
I created a macro program using Python to obtain large amount of image data of each body parts. For each body parts (mouth, nose, ear, eye, hand, foot, knee) I used at least 100 photo data for minimum accuracy. With those image data, I used AI tool provided by Google for machine learning and object identification.
Challenges we ran into
Reaching to minimum accuracy was most challenging. For accurate results, I had to consider the sample size of the data, categories, strategies and more.
Accomplishments that we're proud of
I am proud that I have successfully created a image scrape macro program using Python, considering that I have very little expereince in that area.
What we learned
I have learned that the more data you have, the more accurate result you can get.
What's next for Body Part Analyzer
I want to improve accuracy and categories so that later it can identify human's emotion with visual information.
Log in or sign up for Devpost to join the conversation.