Is AI-powered video search becoming inevitable to security?

Prasanth Aby Thomas | 2018-04-02

Given the increasing affordability of equipment and growing awareness of security requirements, more and more cameras are being installed across the globe every day. While this is a good thing, the sheer volume of footages that come in makes it difficult for operators to find specific objects or people when needed.

This is one area where artificial intelligence (AI) is all set to play a key role. Several security companies are already working on this. Their goal? Make searching through videos as simple as using Google.

According to Matt Sailor, CEO of IC Realtime that recently introduced a video search solution named Ella, this is the future of video surveillance. “Prior to the introduction of Ella and this technology, within the security and surveillance space, finding the proper video clip that matched up with an incident could only be done over a long period of time and by applying search filters and parameters to end up sifting through hours of narrowed down video footage,” he said.

Augment surveillance to save time and money

The advantages of integrating AI-based video search range from efficiency to ease of use. But perhaps a key benefit that customers would find attractive is that it saves time and money. With this technology, end-users across verticals can identify important incidents within minutes.

Stephen Russell, Founder and Chairman of 3VR Security, recently noted that this is the real difference between a traditional video surveillance setup and a searchable video solution. Speaking specifically about the benefit of integrating their solution, Russell had said, “Customers report that they are able to do roughly 10 times the work in one-tenth the time.”

Where do systems integrators benefit?

For security systems integrators (SI), embracing cutting-edge technology is inevitable in pushing their businesses forward. At times, technological advancements do come at a cost for professional installers but integrating video search is not one of them.

Sailor pointed out that his company’s solution offers SIs a direct path to easily attainable RMR. “In installations where a surveillance system does not exist, dealers can easily add the Ella solution without having to swap out the NVRs or surveillance equipment they carry on their trucks and easily add the software solution to the application,” he said.

“On the flipside, Ella can be a powerful tool used by security dealers to revisit past installation with a quick, easy, and powerful add-on service or bring in new business by visiting jobs they may have missed the initial bid on.”

A positive response from end-users

The video search solution is finding takers among end-users across verticals. According to Sailor, this includes residential, commercial, industrial, and traffic applications.

“Since launch earlier this year, we’ve seen a lot of interest from law enforcement who see Ella as a powerful tool that could make their jobs easier,” he added. “With older systems, even if you find the footage you’re looking for, sharing that video with local authorities or outside sources has historically been an arduous task requiring the download and management of proprietary video players. The Ella system generates a tokenized link that’s viewable via any browser and only allows access to the singular video clip you’ve shared that token with.”

Beyond video search to a self-learning solution

While video search is becoming critical to surveillance solutions, the advantages of this solution go beyond this. The power of AI enables software to learn what is important and what is not. It is not enough that operators find certain objects within a video and the system is able to pull up similar objects from other videos. To be truly the Google of video surveillance footage, the solution should be intelligent enough to pick up key incidents by itself.

“Beyond the ability to search video content for exactly what you’re looking for, an additional feature that’s just important is the system’s ability to learn over time,” Sailor noted. “Not only can these systems understand, ‘that is a car’ and ‘that is a person’ they start to recognize patterns and make decisions on which footage is relevant or irrelevant based on that information.”

Bottom Line: yes, video search has become crucial

That intelligent video search solutions will become an integral part of video surveillance solutions is certain. But given the number of companies that are entering the market with this, the question for systems integrators and end-users is which among them provides the best bang for the buck. Advancements in the field of AI and deep learning neural networks are going to push this segment further ahead and it may not be too long before we see more intelligent solutions that can identify incidents, people, and objects even more accurately.


Tags: AI, surveillance, Artificial Intelligence, video search

Go Back