[Updated] This video analytics tool can help retailers profile their customers

SingTel, with KAI Square, launches VAaaS, a video analytics tool for Singaporean retailers, to help them identify customers, as well as count them, among other things


UPDATE: Kai Square today announced that it has secured S$4M (US$3.2 million) in Series B funding from Taiwan-based Ingrasys Technology, a subsidiary of Foxconn Technology Group, and SingTel Innov8. The funds will be used to fuel product development of Kai Square and expand out of Singapore. 

Is there a more efficient and affordable way to distill millions of frames of video footage into meaningful market insights? SingTel aims to answer that question for retailers in Singapore with VAaaS.

SingTel is partnering with KAI Square to launch a VAaaS (Video-Analytics-as-a-Solution) Unified Platform product. It comprises four IP cameras, which would transfer images to the cloud for storage. Business owners will be able to watch surveillance videos and receive event notifications from their mobile or tablet, and the web platform will allow business users to run analytics on the data.

The partners have a key focus on localisation and needs of retailers of various sizes in Singapore. They are also in discussions with a hospital and have been approached by a club regarding their service. Some of the more popular features among retailers include the customer profiling and people counting features.

The customer profiling feature is about identifying the demographics of the store’s customers through the video frames. KAI Square’s product claims to be able to identify the visitors gender and age, and present it in a easy-to-read manner to the user. This would make it easier for marketers the target the right groups of people. Though it is not a 100 per cent accurate, the product works on deep learning technology, where more data obtained would result in the system being more “smart” and accurate with its analysis. As per the company, in an ideal situation, it would be about 70 to 80 per cent accurate. It  is also able to detect emotions of the user, whether he or she is happy or neutral, but is not considered facial recognition technology in the sense that it can identify the customer.

The other popular feature, people counting, basically counts the number of people that are in the store at any given time. This serves as a means to determine store conversion rate, and can be used to optimise staff allocation in certain more popular areas. Other interesting features to note are the crowd density feature which uses thermal mapping technology to detect which areas have more people congregated at; as well as intrusion detection which allows store owners to define high security zones and alerts them if intruders are detected in those regions.

According to Dr Neo Shi Yong, CEO of KAI, video analytics has been in the market as long as video processing. But in the last five years, there has been big leaps in video analytics. This also explains why more businesses have been pumping money into Big Data.

The product will sell at S$200 (US$160) for a month, packaged for two years. According to Dr Neo, this is less than half of the price of their competitors in the US, products of whom some Singaporean retailers are already using.

KAI Square, a SingTel Innov8 investee company, starting discussions with SingTel in 2011 while looking for potential investors, achieving Series A investment in 2012. However, SingTel is KAI’s domestic partner for the distribution of its product, and the eventual expansion to regional markets would be up to KAI’s jurisdiction.

Lee Han Kheng, Vice President Global Products at SingTel stated, “By applying VAaaS in the retail industry, businesses can design better store layouts, drive sales of popular products and deploy manpower more efficiently. It is a solution which eases the two main concerns retailers face today (in Singapore) — pressures from high upfront costs and manpower shortage.”

The post [Updated] This video analytics tool can help retailers profile their customers appeared first on e27.

from e27 http://ift.tt/1iFhCv6
via IFTTT This post first appeared on e27

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s