Case
Emotions sell. But how do you know what emotions a client experiences when visiting your event, showroom, or store? Yuri Kucherovsky from Rockets deals with computer vision and artificial intelligence in the event industry. The emotion expert told our correspondent in an interview how video analytics is used to evaluate staff performance and increase customer loyalty.
— Yuri, tell us a little about where video analytics telegram china phone number came from and how it developed.
In its original form, video analytics appeared in Russia in the late 1990s. Initially, it was video surveillance for security purposes with subsequent manual processing. When it became widespread, the need for video surveillance operators increased sharply. Demand exceeded supply. Professional requirements for operators decreased. As a result, the quality of their work decreased. With the advent of video recorders specially designed for video surveillance systems, the situation changed slightly. The ability to control stimulated a more responsible attitude to work among operators.
Technological progress has led to the widespread introduction of computer technology. Computer-based video surveillance systems with motion detectors have emerged. It turns out that a motion detector can replace several video surveillance operators.
Analytics technologies also developed. Thanks to computing power that made analysis faster and easier, video analytics gained popularity again. Today, it is widely used in security, marketing, and retail. We were the first to use it in the event industry.
— What is the essence of video analytics?
Just as marketing is a fairly broad concept that includes PR, Internet marketing, and event marketing, video analytics is primarily computer vision, which includes recognition of:
persons,
emotions,
objects, etc.
Video data analysis is a subset of computer vision and artificial intelligence. Scientific research in the field of computer vision and artificial intelligence has been conducted in Russia since the 2000s on the basis of research centers and several large universities.
In Russia, until recently, video analytics algorithms were used primarily to detect events, recognize dangerous objects, and identify individuals in order to ensure security at various sites: protected areas, transport (airports, recognition of license plates for the State Traffic Safety Inspectorate), and at government facilities.
dic.academic.ru
Video analytics has a fairly wide application. For example, tracking unwanted persons in retail. Very often, HoReCa establishments, as well as various trading platforms, face shortages, theft and problems of discrepancies between the goods carried in and actually issued. Due to such dishonest work of personnel, a business can lose over 20% of its profit. Video analytics allows you to significantly reduce this percentage and return lost money to the business.
Don't waste your advertising budget
The call tracking and end-to-end analytics service will show the effectiveness of each source in terms of “from click to sale”.
Get a consultation
In the marketing of commercial departments of retail, there is such an important indicator as OSA (On-Shelf Availability) - the availability of goods on the store shelf at any time. In turn, the situation when the goods are not on the shelf, and you want to buy it, is called OOS (Out-of-Stock). Let's imagine a grocery store with 4000 SKUs, for example, the same dairy group. How to understand when the goods on the shelf are out of stock? Usually, the merchandiser finds out about this only at the end of the day, when checking the store shelves. A story arises when a client comes to the store, and the goods are not there. In general, in the market, due to the lack of the necessary goods on the shelf, stores on average lose about 20% of profit.
The solution is a camera that looks at the product shelf. There is a database for each product name, and machine vision in turn tracks whether the product is on the shelf or not. As soon as the buyer takes, say, the penultimate product from the shelf, a notification is sent to the purchasing department. Thus, video analytics allows for a more rapid response and, as a result, returns lost money.
The most cost-effective ways to use video analytics in retail: monitoring the availability and correct placement of goods on shelves, assessing the number of people in queues and preventing theft in stores. Results of implementing video analytics:
from 2 to 5% growth in turnover due to an increase in the availability of goods on the shelf;
+20% increase in turnover and up to 30% increase in average check when implementing a queue detection system and opening additional cash registers;
+10% increase in turnover due to the implementation of planograms based on video analytics.
A great case is the experiment of X5 Retail Group, which the company conducted based on neural networks and artificial intelligence. It used video analytics and computer vision technologies to control the correctness of the layout and quantity of goods on the shelf, track the number of people in queues, determine the most visited departments in stores, and recognize customers' faces, their gender, age, and mood.
The Skolkovo IT cluster's Intelligence Retail development was tested in five Perekrestok supermarkets in the Moscow region and in the Pyaterochka retail chain. X5 chose the most cost-effective ways to use video analytics: monitoring the availability and correct placement of goods on shelves, assessing the number of people in queues, and preventing theft in stores. In addition, video analytics helped improve the quality of service at checkouts.
During the pilot launch, this system learned to recognize about 1,500 products. It demonstrated an accuracy of 93.7% in recognizing products on shelves. As a result, planogram control (product placement schemes on shelves and in store displays) accelerated dozens of times. The number of people leaving the store without buying anything was reduced by 10%, and store losses decreased by 20%. In addition, with the help of video analytics, it was possible to halve the number of thefts in stores.
Source: https://www.x5.ru/ru/Pages/Media/News/060618.aspx
Video analytics is also used to evaluate staff performance and increase customer loyalty. When a user/client/visitor approaches the staff, video analytics allows us to analyze how well they were served and whether their needs were met. For example, we see from the report that out of 1,000 people who spoke to this manager, 33% experienced negative emotions. This is a lot, something is clearly wrong with this manager.
For example, for one chain of petrol stations we set up a push notification system: the management receives a push notification if the level of negativity at the petrol station goes beyond the norm. This increases the speed of response many times over and, as a result, improves the customer experience.
Other business opportunities:
sales forecasting based on data on the actual flow of visitors/buyers;
business efficiency assessment, conversion rate calculation based on statistical data on site attendance;
linking the employee motivation system to the conversion rate;
analysis of the quality of capacity utilization: retail space, staff performance;
evaluation of the effectiveness of advertising campaigns and investments in PR and marketing based on data on site attendance;
reducing personnel costs, adjusting the number of personnel per shift and the facility's operating schedule in accordance with the intensity of visitor flow.
We work with computer vision and artificial intelligence in the event industry. We use facial and emotion recognition in our work.