Recently, in the 「AGI Application hAIkathon」 competition jointly hosted by Feishu and GeekPark Founder Park, and initiated by 11 advanced model and tool providers, the "AI Visual Inspection" project of DMALL stood out among 117 works with its outstanding technological innovation and practical application value, and won the first prize of the "Multimodal Creativity Award".

As a provider of retail digital intelligence services, DMALL's innovation ability and technical strength in the field of science and technology have been verified once again, indicating that it is at the forefront of the times in the application of artificial intelligence.

After the evolution of the large model technology, the application of AGI in various industries is showing a blooming situation. DMALL's award this time reflects its practical implementation of solving real pain points and difficult problems with AI capabilities in the retail industry.

It is introduced that "AI Visual Inspection" is in the inspection scene of the retail industry. Images of key points are obtained through manual photography and调取摄像头 (fetching camera) methods. Through the understanding of the multimodal large model and according to the inspection task indicators, the inspection results are output, making the inspection results of manual photography more objective. At the same time, the camera monitoring is called to reduce labor costs.

With the support of the multimodal large model, the digitalization degree of the existing system continues to deepen. In addition to reducing the dependence on manual nodes, the degree of automation and intelligence is continuously improved, making the business process more closed-loop, promoting the digital intelligence of enterprises, and ultimately becoming a "new-quality productive force".

The relevant technologies of this "AI Visual Inspection" have been applied by DMALL to multiple solutions for the digital intelligence transformation of physical retail. Using advanced image processing and deep learning algorithms, intelligent inspection of retail stores is realized, becoming a "must-have artifact" for chain retail operations.

Taking the key "required course" of store operation, commodity display, as an example, through real-time monitoring and data analysis, the system can help stores timely discover problems such as out-of-stock shelves and irregular displays, and automatically generate tasks and assign them to employees for processing. Then, the shelf photos provided by employees after completing the tasks are analyzed to finally solve the display problem. This kind of efficiency improvement in commodity display can not only reduce the cost and irregularity of pure manual display and replenishment, but also reduce the sales loss caused by out-of-stock, and further provide accurate optimization suggestions for stores by intelligently analyzing data such as sales and customer behavior to help improve operating efficiency and customer satisfaction.

In addition, the scope of AI inspection can also cover key business links such as the sales floor and the back warehouse, including various elements of "people, goods, and the field". For example, if the queue at the cashier is too long, after obtaining this information through AI visual technology, the system can link with intelligent shift management to调来 (call) cashiers to increase manual cashiering; another example is the business safety such as the unobstructed fire passage and the standardized duty patrol, which can be further improved in规范性 (normativeness) through AI inspection to prevent problems in advance.

Since its establishment in 2015, DMALL has been committed to using a variety of new technologies including AI to help retail enterprises carry out a comprehensive digital intelligence transformation to further reduce costs, improve efficiency, and increase income.

Recently, DMALL has established cooperation with first-class large model AIs at home and abroad, laying a solid industry digital foundation for the landing of AGI in the physical retail industry. In a wide range of retail application scenarios, the accumulated "best practices" in retail are combined with the capabilities of the general large model to further optimize products and services. Currently, it has explored dozens of scenarios such as intelligent customer service, intelligent design, and intelligent loss prevention, launched a series of value-added services, and continuously landed, promoted, and iteratively upgraded to allow physical retail to share the AGI dividend.

Zhang Feng, the president of DMALL, once said that DMALL will focus more on AI in the next step, which can better drive the overall improvement of the efficiency of retail industry enterprises, including better serving consumers. There is more space both domestically and overseas.

With excellent technology, carrying out down-to-earth innovation, and promoting the digital intelligence of the trillion-yuan physical retail, DMALL is in action.