New Jersey AI firm ElectrifAi has published 58 machine-learning (ML) models on Amazon Web Services (AWS) that help data scientists and developers prepare, build, train and deploy ML models quickly.

The expanded offering of pre-built and pre-structured ML models for Amazon Sagemaker include models from the computer vision and image analytics collections.

The computer vision models can work on cameras, such as those mounted to drones that can be used to track people, vehicles and animals, even as the camera and item of interest move rapidly. Audio analytics capabilities can also be added.

Computer vision use cases include:

• Smart bird (drone) for perimeter control
• Smart body camera for law enforcement
• Field-ready image-based skin condition identification
• Enhanced workplace safety
• Identification of a specific person in a crowd
• Animal welfare maintenance for farms

The image analytics models can detect, segment and annotate medical anomalies with precision for Sagemaker, allowing one to track the progress of a disease or the success or failure of a treatment protocol. The model can also identify a wide range of skin conditions almost instantly from a picture taken on a regular smartphone.

Examples of image analytics use cases include:

• Use at entry and exit points
• Deployed anywhere with the aid of portable x-ray devices
• Rapid tracking of individuals
• Objective measurement of the effectiveness of treatment protocols over time
• Helping visualise internal anatomical features, disease identifier markers or other respiratory conditions

“Computer vision on the edge is emerging as a game-changer in AI,” said Jim McGowan, ElectrifAi senior vice president. “We’re excited about how these new models will quickly enable new applications. These models hide all of the complexity behind a straightforward interface, simplifying the process of building all kinds of new tools.”

The ML models in AWS Marketplace are available worldwide.

Founded in 2004, ElectrifAi has about 200 data scientists, software engineers and employees with a record of dealing with more than 2000 customer implementations, mostly for Fortune 500 companies. The company is headquartered in Jersey City, with offices in Shanghai and New Delhi.