The Smart Mirror V 2.0 is based on the first version of this showcase but includes Edge AI for on-device transfer learning. A store owner can now quickly retrain trigger products which are mapped to SAP Marketing Cloud recommendations.
Benefits of Edge AI include: privacy (data stays on device), speed (30x faster with TPU), efficiency (balance power and performance), and offline capabilities.
Store owners can easily setup trigger products that link to recommendations from Marketing Cloud. The Smart Mirror uses a “headless” MobileNet, in which the last layer (which makes the final decision on the 1,000 training classes) has been removed, exposing the output vector of the layer before. The Smart Mirror treats this output vector as a generic descriptor for a given camera image, called an embedding vector. This approach is based on the idea that semantically similar images also give similar embedding vectors. Therefore, to make a classification, the Smart Mirror can simply find the closest embedding vector of something it’s previously seen, and use that to determine what the image is showing now.
Comments are closed.