Awesome Instance Segmentation | Photo Segmentation on Custom Dataset using Detectron2
Eran Feit
Deep learning instance segmentation with detectron2 can enable awesome instance segmentation results for photo segmentation on your custom dataset. segmentation detectron2 is a powerful tool for achieving best panoptic segmentation model performance on image segmentation projects.
In this tutorial we will learn how to build a Detectron2 model based on your own custom dataset .
We will learn to train it on Windows and Linux , and code the inference Python code for testing the model on fresh images
You can download the code here : https://ko-fi.com/s/2119cfd494
You can find more cool Instance segmentation projects and tutorials in this playlist : https://www.youtube.com/playlist?list=PLdkryDe59y4Y24C9LW1AjffKmgGUyaInz
A perfect course for learning modern Computer Vision with deep dive in TensorFlow , Keras and Pytorch . You can find it here : http://bit.ly/3HeDy1V
I also recommend this book, https://amzn.to/44GnlLW : "Make Your Own Neural Network - An In-depth Visual Introduction For Beginners ".
~~~~~~~~~~~~~~~ CONNECT ~~~~~~~~~~~~~~~
ā Buy me a coffee - https://ko-fi.com/eranfeit
š„ļø Email : feitgemel@gmail.com
š¤ Fiverr : https://www.fiverr.com/s/mB3Pbb
š¦ Twitter - https://twitter.com/eran_feit
šø GitHub - https://github.com/feitgemel
šø Instagram - https://www.instagram.com/eran_feit/
ā¶ļø Subscribe - youtube.com/@eranfeit?sub_confirmation=1
š Facebook - https://www.facebook.com/groups/3080601358933585
š Medium - https://medium.com/@feitgemel
~~~~~~~~~~~~~~ SUPPORT ME š~~~~~~~~~~~~~~
š
æ Patreon - https://www.patreon.com/EranFeit
#EranFeit #instancesegmentation #python
00:00 Introduction
00:21 Annotate the images & image discovery
09:45 Train the model on Windows (CPU)
22:35 Train the model on Ubuntu (GPU)
25:49 Test the model
~~~~~~~~~~~~~~ Credits ~~~~~~~~~~~~~
Music by Vincent Rubinetti
Download the music on Bandcamp: https://vincerubinetti.bandcamp.com/album/the-music-of-3blue1brown
Stream the music on Spotify: https://open.spotify.com/album/1dVyjwS8FBqXhRunaG5W5u
...
https://www.youtube.com/watch?v=JbEy4Eefy0Y
151583061 Bytes