Object Detection (or face detection) model using Transfer Learning via mobilenet model

Mukul Jeveriya
2 min readMay 26, 2020


Using transfer learning method, we can simply attach our dataset onto already trained model, which saves both resource and time.We shall be using Mobilenet as it is lightweight in its architecture.

In this article I shall show you an example of using Mobilenet to classify images of 5 celebrity. I will then retrain Mobilenet and employ transfer learning such that it can correctly classify the same input image.

So the overall architecture of the Mobilenet is as follows, having 30 layers with

  1. convolutional layer with stride 2
  2. depthwise layer
  3. pointwise layer that doubles the number of channels
  4. depthwise layer with stride 2
  5. pointwise layer that doubles the number of channels


Task steps

Firstly we import MOBILENET from keras application then further code write like import Dense ,Flatten and further things we need we wrote.below are the few images of my work.

prediction 1
prediction 2

GitHub link

I hope this article helps you THANK YOU !!!