Chooch AI launches mobile SDK for training facial biometrics and other AI vision systems
AI training startup Chooch AI has launched a new mobile software development kit (SDK) to beta to enable developers to build enterprise-grade trainable facial and object identification into mobile applications.
The company says Chooch can quickly be trained to perform as a visual expert in any field, identifying aircraft parts or human faces, or identifying and counting cancer cells, among other uses. Chooch neural networks return metadata about presented images or video, including a person’s identity in security screening use cases, or a product SKU. The software can be trained to accurately identify features from any media, such as web-based video, drone feeds, or medical imagery, according to the announcement.
“Software developers can now integrate Visual AI into applications on mobile phones without piecing together a solution,” says Chooch CEO and Co-founder Emrah Gultekin. “We offer a complete platform, from autonomous labeling to training, and an API that no other company can provide. Before this, you needed to kluge together different deep learning components such as data collection, labeling, neural network selection, etc. before you can even train a model. We’ve bundled these steps into a scalable solution, so you can send visual information into the prediction engine and use the output for your applications in real time.”
Chooch says it enables facial recognition to be trained within seconds, reducing costs for tagging visual data by 80 times and increasing video advertising revenue by 5 times for immediate ROI.
The beta SDK is now available for iOS, with the first 1000 API calls free, each subsequent 1000 calls costing $1, and enterprise pricing on request.