Enhancing video analytics with generative AI

Nurdianah Md Nur
3-min read
In this article:

ST Engineering's AGIL Vision allows searches to be conducted in natural language and supports a range of search permutations.

According to Goldman Sachs, artificial intelligence (AI) could drive a seven per cent increase in global GDP and lift productivity growth by 1.5 percentage points over the decade. Generative AI, such as ChatGPT, is among the types of AI that will contribute to that outcome by streamlining business workflows, automating routine tasks, and giving rise to a new generation of business applications.

One way ST Engineering is helping to realise that is through its newly launched video analytics solution called AGIL Vision.

By using generative AI and optimised machine learning models (that already exist in the market), AGIL Vision allows users to perform object searches within vast video footage quickly and accurately. It can identify specific assets like colours, distinctive clothing, brand marks and observable human behaviours to narrow down and quickly pinpoint critical information.

AGIL Vision allows searches to be conducted in natural/human language and supports a range of search permutations wider than the multiple rules that traditional video analytics solutions can handle.

It is expected to transform the operations of public safety and security, defence and education sectors, and enterprises such as retail environments, facilities management and construction. For instance, it could identify a lost child in a crowded shopping mall based on what he/she is wearing.

To simplify adoption, AGIL Vision is a plug-and-play box, which can be set up on-premise or in the cloud. However, the solution does not rely on cloud computing to process data. This ensures data privacy, minimises network latency, eliminates network bandwidth cost, and ensures it works with limited or no internet connectivity.

“With data storage, data flow and processing occurring locally, the risks associated with external hacks or data breaches during transmission to and from cloud servers are removed. User data remains within their environment, reducing the potential exposure to external threats. This approach also enables users to maintain a higher degree of control over their data, strengthening overall data security measures and mitigating vulnerabilities arising from cloud-based architectures,” states ST Engineering in a note to the media.

“The era of generative AI is already creating boundless innovative applications. It is well recognised that AI will be a key transformer of economies and the future of work and life. We are excited with the innovations and opportunities as we embrace AI in our solutions to co-create new possibilities for our partners and customers to try in a new dynamic world,” says Low Jin Phang, ST Engineering’s president for Digital Systems, at the company’s InnoTech Conference earlier today.

At the same event, Dr. Janil Puthucheary, senior minister of State for Communications and Information, says: “To realise the potential of digital transformation, we must equip our enterprises to manage these disruptions effectively, and turn them into opportunities and competitive advantages.”

“The government has and will continue to invest resources in digital infrastructure, which supports enterprises in navigating digital disruptions effectively. [It] will continue to partner enterprises to adopt digital solutions effectively, and to build a more agile and resilient workforce capable of managing and thriving in disruption."

See Also: