Recent advancements in artificial intelligence and machine learning have hugely contributed to the growth of Image Recognition and Object Detection in retail. While Image Recognition and Object Detection are used interchangeably, these are two different techniques. Image Recognition is the process of analyzing an input image and predicting its category (also called as a class label) from a set of categories. For instance, consider an automatic store checkout scenario. The user displays an SKU in front of a camera that is powered by an Image Recognition software. The software, when trained on all the SKUs present in the store, can predict the SKU shown by the user as one among all the SKUs.

Someone got in touch with us recently asking for some advice on image detection algorithms, so let's see what we can do! They already know what algorithms they want to use, so let's start with those. Hang on no, for the uninitiated, let's start with what even is an image detection algorithm?

Before a classification algorithm can do its magic, we need to train it by showing thousands of examples of cats and backgrounds. Different learning algorithms learn differently, but the general principle is that learning algorithms treat feature vectors as points in higher dimensional space, and try to find planes / surfaces that partition the higher dimensional space in such a way that all examples belonging to the same class are on one side of the plane / surface. To simplify things, let us look at one learning algorithm called Support Vector Machines ( SVM) in some detail. Support Vector Machine ( SVM) is one of the most popular supervised binary classification algorithm. Although the ideas used in SVM have been around since 1963, the current version was proposed in 1995 by Cortes and Vapnik. In the previous step, we learned that the HOG descriptor of an image is a feature vector of length 3780. We can think of this vector as a point in a 3780-dimensional space. Visualizing higher dimensional space is impossible, so let us simplify things a bit and imagine the feature vector was just two dimensional. In our simplified world, we now have 2D points representing the two classes ( e.g.