Want to strip out your background using nothing but a webcam and AI?

Don’t have a green screen but still want a virtual background?

You can do it in no time using Python, OpenCV and the pre-trained BodyPix model!

BodyPix is a pre-trained model that allows you to segment your body from nothing but an image. Now, normally it’s only available through Tensorflow.Js but the tf_bodypix library allows you to leverage it through OpenCV and Python! In this particular video we’ll take it one step further and use it to mask out the background in real time by leveraging your Webcam and OpenCV.

In this video you’ll learn how to:
1. Install BodyPix for Image Segmentation
2. Segment your Body in Real Time using OpenCV and your Webcam
3. Remove Backgrounds using BodyPix
4. Creating a Virtual Background in Real Time
Get the code: [ Ссылка ]

A huge thanks to Data Interview Qs for sponsoring this video!
Data Interview Qs allows you to prepare for your next big Data Science interview and to keep your skills sharp with real life interview questions delivered to your inbox three times a week! You’ll get questions straight to your inbox that allow you to hone your skills across a range of data science concepts around data. Sign up for free here: [ Ссылка ]

Chapters:
0:00 - Start
0:13 - Explainer
1:55 - Tutorial Start
2:32 - Install and Import Dependencies
8:58 - Loading the BodyPix Model
11:46 - Real Time Webcam Feed with OpenCv
17:47 - Body Segmentation in Real Time
21:30 - Loading an Image for a Virtual background
26:21 - Applying a Virtual Background
33:00 - Real Time Virtual Background
34:17 - Wrap Up

Oh, and don't forget to connect with me!
LinkedIn: [ Ссылка ]
Facebook: [ Ссылка ]
GitHub: [ Ссылка ]
Patreon: [ Ссылка ]
Join the Discussion on Discord: [ Ссылка ]

Happy coding!
Nick

P.s. Let me know how you go and drop a comment if you need a hand!

свернуть