Скачать или смотреть SynthetikAI | 3D object recognition, segmentation, annotation, and synthetic data generation

DHS has completed an $800K contract award for deployment of SynthetikAI - an automated AI-based platform for 3D object recognition, segmentation, annotation, and synthetic data generation.

The SynthetikAI platform provides a machine-assisted annotation capability to greatly increase the speed, accuracy, and data diversity of 3D CT datasets and to reduce the cognitive burden on annotators and TSOs (Transportation Security Officers).

The DRUID (Digital Removal & Unpacking of Image Data) functionality enables digital unpacking of 3D CT scans into individual components/items and allows users to visualize, identify, classify, and determine the appropriate security outcome for each item. The digitally unpacked items generated by the DRUID functionality can subsequently be utilized to generate completely new synthetic bags by repacking the individual objects with a physics-based packing methodology.

SynthetikAI incorporates ‘on-the-fly’ machine learning such that as new unknown items are scanned, identified, and classified, the system will ‘learn’ the new item, thus increasing system accuracy and reducing false positives over time.

“Synthetik’s solution proposes an advancement to three-dimensional (3D) object recognition that could greatly enhance security checkpoint operations,” said Melissa Oh, S&T’s Silicon Valley Innovation Program (SVIP) Managing Director. “We are excited to see Synthetik’s proof-of-concept—artificial intelligence has many applications to DHS use-cases that we have not yet fully explored.”

This is the fourth award under SVIP’s program for object recognition and adaptive algorithms for airport passenger property screening

SVIP is one of S&T’s programs and tools to fund innovation and work with private sector partners to advance homeland security solutions. Companies participating in SVIP are eligible for up to $800,000 of non-dilutive funding to develop and adapt commercial technologies for homeland security use cases.