Sanjana Roshan
Sep 19, 2022
Introduction
Today retail video analytics has gone past the conventional area of security by giving retailers smart business insights. This data after proper computation can improve customer experience and enhance store performance. This will in turn reduce operational costs and drastically increase the overall productivity.
As explained in [1], high-resolution video cameras on unmanned aerial vehicles are beneficial for applications like smart farming and public safety, among others. Degraded video quality and wasteful resource utilisation might come from inefficient settings in drone video analytics applications caused by edge network misconfigurations. In order to support decision-making for the selection of both network protocols and video qualities in the drones, various supervised and unsupervised machine learning techniques can be of great use.
To compete with upcoming brands and trends, there needs to be a customer-centric approach across upstream and downstream activities in the retail value chain. As described in paper [2], presence of agility, adopting relevant technology and performance metrics to analyse the video surveillance can greatly help in monetary and non-monetary firm performance. Analytical insights from multiple factors like workforce, inventory, customer segmentation, customer value, and customer journey can help organised retailers in reaching higher levels of performance.
Measuring Retail Productivity During the Holiday Shopping Season
Retailers cut costs on well known items and spend vigorously on publicising to draw customers into their stores during the Christmas season. The observations were studied [3] on how customers shopped during the whole Christmas season - from prior to Thanksgiving through Christmas. The video surveillance was analysed utilising a blend of computer vision programming and enhanced by human judgement. The discoveries recommend multiple ways through which retailers could give a more useful shopping experience:
- Customers who showed up sooner than expected on Black Friday had to stand by in a dim parking area for the store to open. This had a negative influence on their insights. Retailers could keep customers in a positive mind-set by giving hot espresso and umbrellas thereby making the experience more enjoyable.
- To manage human resources better, retailers could give pre printed materials having answers to questions that were asked frequently by the customers. This way the staff could attend the customers with unique queries.
- Retailers should consider extending the period for special offers during greater volume of incoming customers. This way customers will be encouraged to browse and buy more products.
- Long checkout lines were an impediment for clients exploring through the store and a burden for customers endeavouring to pay for their buys. Retailers could accelerate the checkout process by improving on complex exchanges and opening more registers.
As discussed in paper [4], Deep Learning approaches (‘multi-layered perceptron networks (MPN)’, ‘Multimedia Tools and Applications convolutional neural networks (CNN)’, ‘recurrent neural networks (RNN)’ and their variants) have played a critical role over the years in video surveillance and its application areas. The growth of video surveillance propels advancements of the video technologies to deal with huge sums of information productively.
Existing research suggests that big data and advanced technologies play an important role for retailers in understanding the integral nuances of their customers. This enhances the customer experience, satisfaction and loyalty. By utilising the video surveillance technology, in-store movements of the customers can be captured and analysed further. The analytics include facial recognition, motion detection and data visualisation techniques which can be used for safety purposes, traffic monitoring, client insights and heat maps.
References:
- Qu, Chengyi, Rounak Singh, Alicia Esquivel-Morel, and Prasad Calyam. "Learning-based Multi-Drone Network Edge Orchestration for Video Analytics." In IEEE INFOCOM 2022-IEEE Conference on Computer Communications, pp. 1219-1228. IEEE, 2022.
- Gupta, Shaphali, and Divya Ramachandran. "Emerging market retail: transitioning from a product-centric to a customer-centric approach." Journal of Retailing 97, no. 4 (2021): 597-620.
- Burke, Raymond R. "The third wave of marketing intelligence." In Retailing in the 21st Century, pp. 159-171. Springer, Berlin, Heidelberg, 2010.
- Subudhi, Badri Narayan, Deepak Kumar Rout, and Ashish Ghosh. "Big data analytics for video surveillance." Multimedia Tools and Applications 78, no. 18 (2019): 26129-26162.