Deep Learning! Easy way to do Image Segmentation with Python

preview_player
Показать описание
Learn how to do image and video segmentation using a state of the art deep learning model. We will use a pre-trained deep neural network that we are going to adapt to our needs to accomplish the task. The code in this tutorial is written in python and the used libraries are tensorflow, keras and opencv.
Image segmentation is a technique from the field of computer vision where we try to predict the exact contour of an object. Deep neural networks are the state of the art solution for tasks like that one and are widely used in the data science/data analytics community.

_About the channel_____________________
TL;DR
Awesome Data science without much math!
--
Hello, I'm Johannes Frey, or simply call me Jo!
On my channel, I will show you how to make awesome things with Data Science. Further, I will present to you some short Videos covering the basic fundamentals of Machine Learning and Data Science like Feature Tuning, Over/Undersampling, Overfitting, ... with Python.

All videos will be simple to follow and I'll try to reduce the complicated mathematical stuff to a minimum because I believe that you don't need to know how a CPU works to be able to operate a PC...

_Equipment _____________________

Рекомендации по теме
Комментарии
Автор

Excellent video. It makes image segmentation so much more understandable. Keep it up!

tntrela
Автор

Thank you so much Joh, your imbalanced learning video was very helpful for my thesis 🙏

kevsersahinbas
Автор

Excellent tutorial on Image Segmentation.

BiranchiNarayanNayak
Автор

Good job! Thank you for the vídeo. A vídeo about Mask RCNN would be awesome!

luisaheise
Автор

Am i right in thinking that the process goes like this: preprocess the data and ensure it can work within the model; train/test the model to ensure it works for the data (improving if not suitable yet); segment/decode the segment map (different approach depending on the model used. i.e. 'fcn_resnet101' pretrained is different to a custom u-net?); present the image with overlay or whatever.

joepearson
Автор

How much faster is inference on the model that just segments people as compared to the existing model that segments all the different classes?

seanfurey
Автор

Thanks a lot man! loved your content and your style. cheers

abhinavmishra
Автор

Sir can you also briefly explain how can we segment hair?

Iceburg
Автор

dude all of this okey but wait. how can we run this on locak machine(windows 10)?

tipveyapayogrenme
Автор

How do I join your company for an internship? Im a fan now!

arnobchowdhury
Автор

Thank you for this amazing lesson. It really helped me a lot with image segmentation. I would like to ask you if tensorflow-gpu is necessary for this kind of tasks or i can just use tensorflow-cpu? Thank you in advance

liakouras
Автор

Please make a video on blend masking pleaseee

Still_i_Rise
Автор

Thank you for your video. Just a quick question. There is a possibility to apply segmentation to pills 💊 ?

Renardbardhi
Автор

My project is on image segmentation of bacteria using rcnn..sir

rajinderrajput
Автор

You sound like Ragnar from Vikings cool....

christyzachariah
Автор

Thanks Johannes for this video, super informative. Are you interested to be a guest on our channel we bring interesting guests such as yourself to talk all things, data !!

learnanalyticsorg