Usually, the machine learning models have to work on large data sets, but the Amazon Rekognition Custom Labels has the ability to work with a limited data set to teach the algorithm a set of objects particular to a given use case
Fremont, CA: Amazon Web Services (AWS) announced its new feature of Amazon Rekognition, called Amazon Rekognition Custom Labels, to help customers build their own specialized machine learning (ML) models, which will help them to understand a set of objects or detect unique objects or scenes, when there is a limited set of information.
Usually, the machine learning models have to work on large data sets to learn something like what's a picture of a dog, as compared to some other animals. Unlike this, the Amazon Rekognition Custom Labels has the ability to work with a limited data set to teach the algorithm a set of objects particular to a given use case.
"Instead of having to train a model from scratch, which requires specialized machine learning expertise and millions of high-quality labeled images, customers can now use Amazon Rekognition Custom Labels to achieve state-of-the-art performance for their unique image analysis needs," the company wrote in a blog post announcing the new feature.
Let's say, if one wants to teach the model to identify a set of engine parts in an auto repair shop's inventory, which usually is a limited set of information, Amazon Rekognition Label detection helps to analyze the objects or scenes and sort the machine parts in the inventory. Amazon Rekognition Custom Labels can help the customers to train their own custom model to identify specific machine parts, such as turbocharger, torque converter, etc.
To start using this, customers can collect as few as ten images for each specific machine part, upload and label those images, and they are guided through each step of the process within the console. Once the dataset is ready fully labeled, customers can use Amazon Rekognition Custom labels with one click, and it automatically chooses the most effective machine learning techniques for each use.
Most machine learning models usually face problems to complete the process with such limited information, but this feature has been specifically designed to learn from a smaller amount of data.
Acting as a publicity agent for Amazon, ACLU, and some shareholders have helped to sell Amazon Rekognition to law enforcement, to help identify faces. This feature offers a more gentle use of similar technology.
Check out: Top Amazon Cloud Solution Companies