Self-checkout device that detects motion in video frames to register products present in the video

A method implemented by a self-checkout device, the method comprising:

• receiving, by a processor, a video footage of a detection plate of the self-checkout device captured by one or more cameras, wherein the detection plate adapted to allow placement of products thereon, and
• wherein the one or more cameras are positioned to have a Field-of-View encompassing at least the detection plate;
• detecting, by the processor, a presence of motion in the video footage by processing thereof;
• selecting, by the processor, a sequence of video frames over a time interval corresponding to the detection of the presence of motion in the video footage;
• registering, by the processor, one or more products present in the sequence of video frames by:
• analyzing, by the processor, the sequence of video frames to detect one or more objects therein;
• isolating, by the processor, the detected one or more objects in the sequence of video frames and to extract visual features of the detected one or more objects;
• converting, by the processor, the extracted visual features of the detected one or more objects into an embedded feature vector;
• comparing, by the processor, the embedded feature vector with pre-stored feature vectors in an embedding database, and
• identifying, by the processor, the detected one or more objects based on the comparison,
• wherein the identified one or more objects are registered as the one or more products;
• fetching, by the processor, prices of the registered one or more products;
• generating, by the processor, a total bill based on the fetched prices and processing a payment for the total bill;
• and
• displaying, by the processor, the total bill on a display screen of the self-checkout device.

A non-transitory computer-readable medium having machine-readable instructions stored therein, which when executed by one or more processing units, cause the one or more processing units to perform the steps comprising:

• receiving, from one or more cameras, a video footage of a detection plate of the self-checkout device,
• wherein the detection plate adapted to allow placement of products thereon, and
• wherein the one or more cameras are positioned to have a Field-of-View encompassing at least the detection plate;
• detecting presence of motion in the video footage;
• selecting a sequence of video frames over a time interval corresponding to the detection of the presence of motion in the video footage;
• registering one or more products present in the sequence of video frames, by:
• analyzing the sequence of video frames to detect one or more objects therein;
• isolating the detected one or more objects in the sequence of video frames and to extract visual features of the detected one or more objects;
• converting the extracted visual features of the detected one or more objects into an embedded feature vector;
• and
• comparing the embedded feature vector with pre-stored feature vectors in an embedding database, and to identify the detected one or more objects based on the comparison,
• wherein the identified one or more objects are registered as the one or more products;
• fetching prices of the registered one or more products;
• and
• generating a total bill based on the fetched prices and processing a payment for the total bill.