Shelf Image Recognition System for Product Detection and Counting

The objective of this project is to develop a robust image recognition system tailored for processing images of shop shelves in batch. The system must employ state of the art deep learning techniques to accurately detect and count all products displayed on the shelves. Additionally, it will be deployed as a microservice in the cloud, providing endpoints for uploading images of products to detect and images of shelves, and retrieving results including the positions of detected objects and their categories.