Landmarks Recognition Web-App using Python | Like Google Lens | Machine Learning Projects

Hello readers, in this blog we are going to learn about how we can create Landmark Recognition Web-application using Python, Streamlit, and Tensorflow.

We need to divide the work into separate parts.

  • Detect the Landmark using the trained model

  • Find the Longitude, Lattitude, and address of the detected landmark

  • Display all the results in application UI


(1) Detect the Landmark

Creating the model for detecting landmarks is not an easy thing, we need to have powerful processing, and there are so many try n errors are there for training a perfect model. So in our application, we are going to use a landmarks_classifier_asia_V1 model. In this model, there are 98959 classes available. This model is only limited to Asia's landmarks only.


Implementation

  • Download the Model from here.

  • Download Labels from here.

  • Follow this code.

Code

Result



(2) Find the Address, Latitude & Longitude of detected Landmark

In this part we are going to fetch the address, Latitude, and Longitude information, so later on we can plot the landmark on the Map in our web application.


Code & Output

(3) Create Web-application using Streamlit

We are almost going to finish the project, out two main tasks is done, now we need to create the user interface of our application, then we need to merge our code with the UI of Streamlit.