When examining a product in the store, users can use augmented reality (AR) devices to research the product. However, unlike in online shopping, making product comparisons while shopping in the store is time-consuming, burdensome, and error prone. This disclosure describes the use of object recognition techniques to enable AR-based visual searches and product comparisons. With user permission, products in the user’s field of view are detected and identified, and a dynamic buffer of product information is maintained. The buffer serves to transform the searches for individual products within a visual search session into a ranked list that can be sorted and compared based on any number of criteria and facilitates seamless comparison of products when shopping in a store.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Sharifi, Matthew and Sharma, Abhanshu, "Scanning and Ranking a Stream of Visual Objects", Technical Disclosure Commons, (June 18, 2021)