Why

API uses ML to detect free (not occupied) parking places on your image.
Provide frame from camera stream, photo, etc - get count of free/busy places.
Use Api as part of parking monitoring automation, Bot with informing and other solutions.

• I’m using tg bot to find not occupied places at parking near an apartment building.

Try different ML models, select best for your application.

Feature request

Free free to ask any questions and request features.
Also it’s possible to train model by your images to improve recognition results.

Some usefull info

• /detectfile - endpoint accept file as form data (‘content-type: multipart/form-data’)
• /detect- endpoint accept file as base64 string (‘content-type: application/json’) - plain text or data_uri

  • plain base64 text, like /9j/4AAQSkZJRgABAQ...

  • data-uri with base64 text, like image/jpeg;base64,/9j/4AAQSkZJRgABAQ...

  • ‘bytes to string’ like b'/9j/4AAQSkZJRgABAQ...

• /models - endpoint returns availiable models list. n1s_07042023 - (s)mall model trained on dataset 1 at date ddmmyyyy