2019 Trends – Interview with our Product Manager
 |
||
|
With increasingly rapid advancements in artificial intelligence (AI) and machine learning software, the video surveillance market is continuously evolving. We had a chat with VSP’s product manager, Andrew Vo-Nguyen, to gain expert insight into the present AI industry and trends that are expected to impact the video surveillance market in 2019 and beyond.
|
||
‘Artificial Intelligence’ is an increasingly hyped field, yet seemingly lacking in the current number of technology deployments. Do you think that will change in 2019?
|
||
|
Besides AI, what technology trend do you think will have the greatest impact on the market in 2019?
Another trend in surveillance technology is diversifying the use of CCTV footage in different verticals. Combined with analytics, CCTV cameras can be used for other purposes, such as retail marketing, traffic control, parcel tracking, and crowd management. The increased flexibility of these devices means that the cost of a surveillance system can be justified through its coverage across multiple sectors of a business. This is making CCTV more affordable than ever.
|
||
What are the known challenges facing AI implementation?
One common challenge facing AI are the large costs involved in developing complex algorithms to help reduce false alarms and to correctly identify targets. The computing power involved with deep learning intensifies as camera resolutions continue to rise. Software packages can take weeks before a working model of the scene can be used. Hardware (especially GPUs) will need to be more affordable, and thus accessible to a greater clientele, in order for AI to really take off.
|
||
|
With the technological developments in analytics and cloud-based video solutions advancing rapidly, how do you see video management software evolving moving forward?Analytics engines already exist in cameras such as Axis and Hikvision, however the algorithms are static and lack deep learning capabilities. Moving forward, deep learning analytics will be processed on the edge. As GPU hardware gets more powerful in smaller form factors, cameras will run analytics independent of processing servers. These devices will only transmit the metadata of the analytics events back to a central database where it can be compiled into a presentable manner for the end user. |
||
|
What are your expectations for the evolution of the Video Surveillance-as-a-Service (VSaaS) market in 2019 and beyond?Cloud-based video and analytics work in tandem to provide the end user with a user-friendly experience similar to subscription-based cloud services already in use on their mobile devices. There will be a point where companies that currently supply internet, streaming and telecom services will take on surveillance and video analytics. Proof-of-concept sites utilising 5G are already being rolled out to trial the use of cloud-based video with analytics. Increased accessibility to surveillance services will mean a busy future for the VSaaS industry! As such, VSP will continue to embrace and evolve with the rapid rate of technological developments.
|
||
|
Interview questions inspired by securityinfowatch.com |