Background: Some early applications of Computer Vision in Retail arose from e-commerce use cases - but increasingly, it is being used in physical stores in a variety of new and exciting ways, such as:
● Optimizing merchandising execution, in-stocks and sell-thru
● Enhancing operational efficiencies, enable real-time customer engagement
● Enhancing loss prevention capabilities, response time
● Creating frictionless experiences for shoppers
Abstract: This talk will cover the use of Computer Vision in Retail, the implications to the broader Consumer Goods industry and share business drivers, use cases and benefits that are unfolding as an integral component in the remaking of an age-old industry.
We will also take a ‘peek under the hood’ of Computer Vision and Deep Learning, sharing technology design principles and skill set profiles to consider before starting your CV journey.
Deep learning has matured considerably in the past few years to produce human or superhuman abilities in a variety of computer vision paradigms. We will discuss ways to recognize these paradigms in retail settings, collect and organize data to create actionable outcomes with the new insights and applications that deep learning enables.
We will cover the basics of object detection, then move into the advanced processing of images describing the possible ways that a retail store of the near future could operate. Identifying various storefront situations by having a deep learning system attached to a camera stream. Such things as; identifying item stocks on shelves, a shelf in need of organization, or perhaps a wandering customer in need of assistance.
We will also cover how to use a computer vision system to automatically track customer purchases to enable a streamlined checkout process, and how deep learning can power plausible wardrobe suggestions based on what a customer is currently wearing or purchasing.
Finally, we will cover the various technologies that are powering these applications today. Deep learning tools for research and development. Production tools to distribute that intelligence to an entire inventory of all the cameras situation around a retail location. Tools for exploring and understanding the new data streams produced by the computer vision systems.
By the end of this talk, attendees should understand the impact Computer Vision and Deep Learning are having in Consumer Goods, key use cases, techniques and key considerations leaders are exploring and implementing today.
Speaker
Brent Biddulph, GM, Retail & CG Solutions
Hortonworks
Hello, and thanks for joining this session titled: Computer Vision: Coming to a Store Near You.
My name is Brent Biddulph, GM, Retail & CG, Cloudera
Joining me is Florian Mullerklein, Data Scientist, Miner & Kasch
AND INDEED, it likely CV may ALREADY be in a store near you.
Today’s talk we will share how innovation using computer vision is already impacting brick & mortar…
…illustrative use cases retailers are testing and deploying today.
And finally, sharing foundational concepts of how it works, and what it looks like.
So why are we talking about “in-store” when all the press lately has been about traditional retailers improving their online / omni-channel capabilities?
For many - focus and attention is shifting back to the core business – brick and mortar - where more than 86% of top line revenues still occur.
New technologies enable new possibilities, a few of which we will share today.
IoT is not just reserved for industrial based industries.
Retailers are also leveraging new technologies to gather “more observations” and generate “real-time responses” to improve customer experiences and operational efficiencies.
Video streams are an emergent area showing much promise in retail (we will look at a few examples shortly)
Applying Computer Vision techniques can be applied to tackle Key Business Imperatives such as:
Enabling Frictionless Commerce – customers are coming to EXPECT more and more as part of overall CX
Improving Operational Efficiencies – providing insights into operational challenges such as OOS (-5% revenues) – 24% goes to Amazon
Improving Customer Experiences – providing image based search capabilities (50% of buyer search inspiration)
Reduce Fraud & Shrink – on sales floor, checkout, receiving dock, fresh foods departments (-5% of revenues)
Perhaps the most widely known example in NA, was the public launch of the Amazon Go store summer of 2018.
A convenience store at 1,800 sq. ft.
This ‘test’ originally ran 2 years in incubation, then rolled out to employees for a full year before opened to the public.
Today, Amazon has since opened additional locations in Chicago, San Francisco with plans to add more, with larger assortments in NA as part of their brick and mortar strategy.
Alibaba has now opened 100 HEMA retail stores in the past year, in an effort to merge online and offline retail.
- Customers use an app to scan products, get information (country of origin, recipes, etc.) and pay for their groceries.
- The stores double as distribution centers, where employees fill bags with online orders, then place them on a conveyor belt to the delivery center.
- Customers pay via Alipay. Customers can even pay by scanning their faces at kiosks.
Perhaps one of the most promising high value uses cases for CV ‘in-store’ is tackling the age-old challenge of merchandising execution and in-stocks.
Leading grocers like Walmart, Kroger, Ahold have been testing for over a year…
…and Ahold-Delhaize being first to announce a chain wide rollout across 500 Giant Store locations US East Coast.
These robots can scan a 40’ aisle in 60 seconds, scan an entire store 3x day, improving accuracy over time.
ESLs now available with built-in cameras enable ‘cross aisle’ views, similar to the mobile robots
Some retailers are using in-store cameras pointed at the entry to identify ‘guest engagement opportunities for Customer Service associates.
In fashion and beauty, “magic mirrors’ can improve the CX at stores like Nieman Marcus and Sephora > whether in-store or at home on your own mobile device.
Allowing for speedy assessment of style, color and ‘looks’.
Disrupters like Stitch Fix and Untuck It embedded this as a core capability.
The possibilities for CV to impact retail seem limitless.
Extending online analytic capabilities (and more) to the physical brick & mortar rhealm are gaining traction.
Smart Carts - image recognition, payment + scales – sensor fusion (caper)
Ceiling mounted cameras - In-store customer paths, dwell time, even distinguishing staff from customers (improve engagement, CX, hot spots, choke points, conversion, etc.)
Robots & ESLs – with built in cameras (facilitate proximity marketing, dynamic pricing and pick-pack pathing at the item, store level) – AND provide merchandising execution insights much like the mobile robots.
Satellite images capture retail and mall parking lot traffic (orbital insights). Competitive insights for retailers, new store opening, traffic counts, construction impact, etc.
Autonomous vehicles – e.g. Stop & Shop (Ahold Delhaize USA) testing delivery AND mobile stores (Kroger, Walmart also continuing to test in pilot markets)
Now the legacy Clouderans especially may think of ML on Cloudera as simply a CDSW play.. but here’s what industrialized ML actually looks like on the platform:
First, data streaming or batch ingested into the pipeline
Data engineering to clean and prep
Building and deploying models
Scoring and feeding results through to BI dashboards or real-time data views
The bottom line is that Enterprise ML requires the big picture.