Video analytics has emerged as one of the most popular applications of computer vision and is being widely used in surveillance within the retail industry. BriefCam, a company that used computer vision-based compression technology to generate analytics, was recently acquired by Canon. The company not only provided video compression for storage, but also enabled the rapid generation of analytics.
Tractica recently sat down with the company CEO, Trevor Matz, to discuss the acquisition, technology, and business. Excerpts from the dialogue are below.
Congratulations on your acquisition by Canon. How will the acquisition help you position and grow your business?
We believe that the acquisition will drive rapid innovation in video analytics by BriefCam, as well new co-innovation activities with Canon and its portfolio companies. The acquisition will allow BriefCam to enter new markets, deliver stronger vertical solutions, and serve its global customer base effectively.
What are the key aspects of your technology that differentiate you?
BriefCam is a complete video content analytics platform, which provides industry-leading video synopsis and deep learning solutions. Our customers have been using these solutions for rapid video review and search, real-time alerting, and quantitative video insights. Our technology transforms raw video into actionable intelligence by detecting and tracking all objects in a frame, extracting those objects, and classifying them for rapid review and search, quantitative insights, and proactive response. We are one of the very few companies that enables the fusion of computer vision, deep learning, and business intelligence technologies.
Our value add comes from a platform that provides an easy, seamless way to address a range of safety, security, and operational efficiency use cases for any organization.
What is your monetization strategy?
BriefCam‘s software is currently available both via a subscription and a perpetual licensing model. Per camera pricing is set to enable full camera coverage in any size. In the law enforcement sector, customers can take advantage of file-based video ingestion, in addition to video management system (VMS)-based ingestion.
In which industries are you seeing traction today and in any particular use case?
About 60% of BriefCam‘s business is in the law enforcement and public safety sectors. The remainder falls within key verticals such as transportation agencies, major enterprises, healthcare, and educational institutions. The BriefCam platform is used both for safety and security use cases, as well as operational efficiencies use cases such as:
- Tracking crowd demographics, size, and movement patterns
- Identifying crime hotspots
- Designing more profitable retail floor plans
- Uncovering optimal store locations
- Optimizing traffic flow at major interchanges
- Tracking employee compliance with safety regulations
How has your revenue growth been during the last three years and the last year?
BriefCam achieved 100% growth year-over-year in 2017 and is targeting 100% growth in 2018.
What are some of the key challenges you had to overcome over the years?
The market is transitioning toward the IP-based camera systems. This has opened up significant opportunity for integrators and resellers to deliver value-added solutions to customers. With the presence of IP cameras, BriefCam is able to deliver video content analytics solutions that transform petabytes of video into valuable intelligence for an organization. In addition, the demand has been strong for analytics in the video surveillance market spurred by today’s terrorist-related events.
In your opinion, what are some of the challenges pertaining to hardware when it comes to running video analytics?
BriefCam can be configured to process video on-demand or in real time. Small systems, with up to double digit numbers, can be run on a high-end Xeon system. Systems that have tens of thousands of camera streams need graphics processing units (GPUs) for processing. BriefCam collaborates with NVIDIA’s GPU technology to accelerate the processing capabilities and leave it to customers to choose the model and associated hardware that meets their use cases. We have been innovating the software to offer faster processing ability that enables real-time processing of camera feeds all the time, in a cost-effective manner.
Do you see any movement toward the edge?
We definitely see a need for edge processing in terms of distributed architectures where video is processed at the “edge,” e.g., at the retail store level, and not only centrally at headquarters where the main VMS resides. On-camera analytics is not something we have on our short-term roadmap, but we do keep a very close eye on the evolution of camera technologies and their capabilities, and evaluate our go-to-market strategy accordingly.