Intel joins hands with hanshuo and Microsoft to release the "Ai + retail" trick!

In the increasingly competitive retail market ,AI、 Computer vision 、 The Internet of things And other digital innovative technologies are reshaping the core competitiveness of retailers . Not only help retailers reduce operating costs 、 Improve operational efficiency , Also by creating convenience 、 Personalized offline shopping experience , To win the favor of consumers , So as to obtain higher business growth . therefore ,“AI + retail ” It has become the retail industry to maintain vitality in the post epidemic era 、 The inevitable exploration direction of the new machine .

In this context , Intel and Hanshuo 、 Microsoft Teamed up with , Around retail scenarios AI Infrastructure innovation 、AI Application optimization and other aspects of close cooperation , Launched Based on Intel   Digital store and self-service cashier loss prevention scheme based on the architecture .

This solution integrates professional digital store solution provider hanshuo in “AI + retail ” Long term technological innovation and industry best practices in the field , Intel's full stack technology optimization from software to hardware , And Microsoft's AI And Internet of things Technology , Help retail customers build high performance 、 High insight 、 High economy 、 An easy to implement smart retail management system , Solve shelf commodity management 、 Self service cashier 、 Thorny problems in commodity loss prevention and other links , Create greater value for global retail customers .

Global retail AI The road to explore

Now ,AI The important value displayed has been widely concerned by retailers all over the world . Through the application in retail business AI technology , Retailers can gain more effective insight into consumer preferences , Provide consumers with personalized 、 Unique service , Enhance the attractiveness of retail services . meanwhile ,AI Help retailers turn more processes into automated processes , And accelerate to gain more insight from massive retail data , And realize more benefits . data display , To 2035 year ,AI It will increase the profit margin of retail and wholesale businesses by nearly 60%1.

At present , With the overall restoration of retail order , Retailers use more unmanned retailers 、 Retail robots , And support AI Real time video stream analysis to monitor the work area , Ensure social distance compliance .

                                                                     Retail robot illustration

although AI Help retailers stand out in dynamic markets , however ,AI More in-depth applications are also faced with technology maturity 、 Users' willingness to accept 、 Many obstacles in terms of cost . Lack of skills 、 Fear of the unknown 、 Difficult to find AI The appropriate starting point for the strategy is considered AI The first three obstacles to deployment 2.

To solve the above problems , And enhance retailers' understanding of AI Confidence in application , Intel and hanshuo 、 Microsoft innovates around digital store infrastructure 、 Application optimization 、 Extensive cooperation has been carried out in ecological construction and other aspects , A set of Based on Intel   Framework of hanshuo retail AI The plan came into being . Committed to helping retailers restructure “ people 、 cargo 、 site ” Relationship , Create digital application scenarios , Improve the overall operation and management efficiency of stores .

The three parties work together , Cohesion AI + Retail practice

Intel and hanshuo 、 Microsoft teamed up , Focus on smart shelf management and self-service cashier loss prevention in digital stores , Created innovative solutions ：

AI Digital store , Smart shelf management

Digital stores are a collection of digital technologies and applications in many fields , Among them, shelf management is one of the keys to the construction of digital stores . In order to strengthen shelf management , The scheme is passed by Price tag detection 、 Commodity identification 、 Shortage detection 、 Action behavior detection 、OCR distinguish And so on , Identify the goods in the shelf 、 Classification and counting , So as to automatically determine whether there is a shortage of goods in the shelf 、 Wrong placement and other problems .

be based on AI Hanshuo's digital store program provides fixed cameras 、 Mobile phones 、 Different shelf management options such as robot inspection , It realizes the display collection based on electronic price tag and commodity identification 、 Monitoring and verification . When it is detected that the goods in the display space are in an abnormal state such as out of stock , It can remind retailers of abnormal display in time . meanwhile , On the premise of complying with consumer privacy laws and regulations , The scheme can also realize the identification and tracking of consumer attributes , Count the residence time of consumers 、 Location, etc , Generate the heat map and moving line in the store , Assist in precision marketing . 

                                                                 Digital store map notes

By carrying Intel   core   Processor edge server / Intel   video AI Calculation box , The scheme meets AI The stringent performance requirements of loads such as model reasoning . Intel   core   The processor has excellent performance per watt 、 Rich graphic effects and I/O Integration features , Support the access of various image sensors . The processor also integrates Intel sharp torch  Xe  The graphics card , Able to handle complex AI The workload , You can also use Intel  Media SDK Hardware acceleration , Improve video transcoding 、 The speed of image processing and media workflow .

meanwhile , Thanks to the Microsoft Azure Cognitive Services, This solution can provide fluency 、 Advanced algorithm model , And a highly secure seamless user experience , It can be used without machine learning expertise . meanwhile , Also equipped with Microsoft Azure Machine Learning, Use industry-leading system integration tools and machine learning system integration tools to accelerate the speed of value realization .

be based on AI Independent cashier loss prevention

Although there are scale differences and diversity in the retail industry , But almost all retailers will suffer from the loss of revenue due to the loss of goods . Especially in the self-service cashier business , Retailers will face more severe loss prevention challenges .

                                                                         Self service cashier chart note

The scheme adopts computer vision technology , It can be targeted at missed scanning in real time without feeling 、 Did not pay 、 Identify abnormal shopping behaviors that account for a large proportion, such as price mismatch . Hanshuo passed pedestrian detection 、 Pedestrian recognition 、 Article detection 、 Item search 、 Accumulation of technologies such as multi-target tracking , On the basis of millions of data , Has reached 98% The above algorithm accuracy .

The program is powered by Intel   Processor's self-service radio to run depth vision related loads , And use OpenVINO  Tool suite to accelerate the performance of depth vision applications , So as to fully meet the application needs of self-service cashier loss prevention . Effectively reduce the losses caused by abnormal behaviors in the process of cashier settlement , And human imperceptible abnormal behavior recognition , It also greatly improves the overall shopping experience of users .

Next Station , From retail informatization to smart retail

Intel 、 Hanshuo 、 Microsoft's successful cooperation in retail has promoted AI Application practice in retail industry , Stimulate retail technology innovation , Give full play to AI Technology is mining the value of retail data 、 Optimize retail business processes 、 Important value in improving consumers' shopping experience .

stay AI Performance aspect

Intel   core   processor + OpenVINO  The combination of tool kits can be used without using expensive dedicated hardware , Greatly improve product identification 、 Product tracking 、 Pedestrian identification 、 Pedestrian classification, etc AI The efficiency of the algorithm , Provide high cost performance 、 agile AI Computational support .

In the distribution of computing power

Han Shuo general AI The algorithm converges on Intel based   Processing in the edge server of the architecture , And interact with the cloud management platform , To build the “ cloud - edge - End ” Fusion data processing mode . This scheme gives full play to the improvement of Intel products and technologies AI Advantages in computing power , At the same time, the delay of data processing is reduced by edge processing , Save investment in key resources such as bandwidth .

AI、 Internet of things technology is giving birth to a far-reaching change in the retail industry , By applying digital innovation technology to shelf management 、 Precision marketing 、 Supply chain management 、 Safety management and other fields , Retailers are expected to reshape the traditional retail “ people ”、“ cargo ”、“ site ” The relationship between the three elements , More fully tap the value of massive retail data .

In the innovation process of smart retail solutions , Intel and hanshuo 、 Microsoft will be in AI Infrastructure Optimization 、AI Algorithm innovation 、 Carry out more in-depth cooperation in the construction of smart retail ecosystem , Provide more retailers with better products and solutions , Contribute to the reform of the global retail market .

Data sources ：

[1] Business Insider. THE FUTURE OF RETAIL 2018: ARTIFICIAL INTELLIGENCE.

[2] Gartner. 3 Barriers to AI Adoption.