Computer Vision As A Means to A Smart Retail

Our daily routines and environments have been increasingly integrated with smart technologies for the last couple of years. The real-time feature has altered from what a conventional environment is to a smart environment. The variety of smart solutions and their ease of deployment have gained themselves notable attention, especially in the aspect of surveillance.

Two surveillance problems that can be solved by implementing smart solutions are occupancy prediction and human activity detection and recognition. Occupancy prediction becomes highly important for those businesses who serve spaces as their products. Furthermore, the occupancy prediction becomes significantly needed in times of Covid-19 pandemic and life after it since people would worry when they see crowds.

It is what we call computer vision is the answer to those problems. According to SAS Institute Inc., computer vision is a field of artificial intelligence that trains computers to interpret and understand the visual world. Utilizing the digital images from cameras and videos and processed by deep learning models, a machine can accurately identify and classify objects.

As mentioned before, for businesses whose main product is providing spaces for other parties, computer vision might be the golden way to unleash maximum profit. Calculating about human traffic, determining strategic areas, or simply recognizing some particular objects are only one solution away.

In Indonesia itself, the optimization of computer vision is growing. Computer vision is being considered as solutions that can boost business activities and maximize profits. Envisioning a smart business is no longer just a fantasy.

fX Sudirman was a good example of how smart business operates. Together with Qlue, fX Sudirman had turned the fantasy into reality by making a collaboration under the solution called QlueVision. This collaboration was not only interesting in the sense of technology but also of business. 

QlueVision had enabled fX to scale up their performance by enhancing the existing CCTVs to deliver more insights. One case that was implemented was the ability to count the numbers of people entering and exiting the mall area. This helps retailers to get more accurate occupancy rates in their areas, allowing them to be more precise in complying with the health protocol for Covid-19.

Moreover, QlueVision was demonstrated to be able to analyze the movement of visitors which then visualized into a heatmap. Not only to help retailers define the rental fee, this advantage can go beyond that aspect. Retailers can also identify visitors’ shopping behavior through computer vision.

Other than that, more capabilities served by QlueVision includes facial recognition, mask detection, gender and age identification. All of these capabilities are seamlessly done with up to 98.5% accuracy on the existing CCTVs. The data captured can be measured and visualized automatically through QlueDashboard, an all-in-one platform to make sense of all data recorded by visualizing them so that the whole process will operate seamlessly.

Growing a retail business might be tough. But, growing a retail business, adapting to an emerging pandemic, then facing the new normal after it, might need a transformative solution. It is truly the right time for retailers to start implementing AI technologies.