Data bridge between anonymous browsers and interactive devices
OwnerPatrick Nunallyinventor
Life of this patent
Estimating the term…
Abstract
Embodiments described herein provide systems and methods for correlating anonymous user data with in-person user sensor data. In some embodiments, a user interacts with a web interface which collects sensor data (e.g., video and audio) and stores that sensor data with other customer information (e.g., browsing history). The collected data is sent to a local interaction device which collects its own sensor data, which is compared to the data in the packets to correlate the in-person user with the online user, and continue the customer experience.
Claims14
A method of correlating anonymous customer information, comprising:
- receiving, at a local interaction device, a packet including:
- customer sensor data from at least one of a microphone or a camera, and
- customer interaction information,
- wherein the packet is distributed to the local interaction device, from among a plurality of
- local interaction devices, based on a geographic mapping of one or more locations most likely to be visited by the customer, and wherein the packet includes a retention policy that causes the packet to expire;
- capturing an image of a customer;
- capturing an audio recording of the customer;
- performing a comparison of the image and the audio recording with the customer sensor data, the comparison including performing facial recognition based on the image and subsequently confirming the facial recognition based on the audio recording;
- and
- generating a first embedding of the sensor data from the camera via a neural network;
- generating a second embedding of the image of the customer via the neural network;
- and performing a comparison between the first embedding and the second embedding;
- wherein the comparison includes computing the Euclidean distance between vectors, and performing an action with the customer interaction information based on the comparison.
A system for correlating anonymous customer information, comprising:
- a memory storing processor executable instructions;
- and
- one or more processors that read and execute the processor executable instructions from the memory to perform operations comprising:
- receiving, at a local interaction device, a packet including:
- customer sensor data from at least one of a microphone or a camera, and
- customer interaction information,
- wherein the packet is distributed to the local interaction device, from among a plurality of local interaction devices, based on a geographic mapping of one or more locations most likely to be visited by the customer, and wherein the packet includes a retention policy that causes the packet to expire;
- capturing an image of a customer;
- capturing an audio recording of the customer;
- performing a comparison of the image and the audio recording with the customer sensor data, the comparison including performing facial recognition based on the image and subsequently confirming the facial recognition based on the audio recording;
- and
- generating a first embedding of the sensor data from the camera via a neural network;
- generating a second embedding of the image of the customer via the neural network;
- and
- performing a comparison between the first embedding and the second embedding,
- where in the comparison includes computing the Euclidean distance between vectors, and,
- performing an action with the customer interaction information based on the comparison.
A non-transitory machine-readable medium comprising a plurality of machine-executable instructions which, when executed by one or more processors, are adapted to cause the one or more processors to perform operations comprising:
- receiving, at a local interaction device, a packet including:
- customer sensor data from at least one of a microphone or a camera, and
- customer interaction information,
- wherein the packet is distributed to the local interaction device, from among a plurality of local interaction devices, based on a geographic mapping of one or more locations most likely to be visited by the customer, and wherein the packet includes a retention policy that causes the packet to expire;
- capturing an image of a customer;
- capturing an audio recording of the customer;
- performing a comparison of the image and the audio recording with the customer sensor data, the comparison including performing facial recognition based on the image and subsequently confirming the facial recognition based on the audio recording:
- generating a first embedding of the sensor data from the camera via a neural network;
- generating a second embedding of the image of the customer via the neural network;
- and performing a comparison between the first embedding and the second embedding,
- wherein the comparison includes computing the Euclidean distance between vectors, and
- performing an action with the customer interaction information based on the comparison.
Dependencies are inferred from the claim text.