The customer has asked for a sophisticated solution to address the challenge of accurately detecting and recognizing products on shelves. With operations spanning multiple countries, the client aimed to streamline and enhance their product recognition process.
DataObrii built the labelling framework for multiple purposes. It allowed doing a full cycle of creation, verification, and testing of the training set of labelled images, each including about 200 objects.
To this end, we used:
- TensorBox framework with inception v4 model for regression and classification problems;
- OpenCV for visualization of the results, a training set labelling tasks;
- deployment to the Google ML Engine for the distributed training of the model;
- cascade recognition model to be able to recognize different types of products.