Non-transitory computer-readable recording medium, generation method, and information processing device
An information processing device obtains each piece of image data captured within a period of time from entering until exiting of a person at a store. The information processing device identifies joint positions of a skeleton related to the person by analyzing each piece of the image data. The information processing device identifies, as an action which indicates a degree of interest of the person in the product, an action performed by the person to a product in the store from the entering until the exiting, on a basis of the joint positions of the skeleton. The information processing device generates a detection rule that correlates the identified action and the product with each other.
1. A non-transitory computer-readable recording medium having stored therein a generating program that causes a computer to execute a process comprising:
obtaining each piece of image data captured by a camera installed in a store within a period of time from entering until exiting of a person at a store;
generating joint positions of a three-dimensional skeleton of the person by inputting each piece of the image data into a trained machine learning model;
identifying joint positions of a skeleton related to the person by analyzing the each piece of the image data;
identifying the product on the basis of the each piece of the image data and coordinates of the product in the each piece of the image data and a size of identified product by referring to a product database storing information about products;
determining a degree of interest of the person in the product, on a basis of the identified action for every size of the product;
identifying, as an action which indicates the degree of interest of the person in the product, the action performed by the person to the product in the store from the entering until the exiting for every size of the product, on a basis of the joint positions of the skeleton;
and generating a detection rule that correlates the identified action and size of the product and the determined degree of interest.
2. The non-transitory computer-readable recording medium according to claim 1 , wherein the process further includes judging a posture of parts in the person based on the joint positions of the three-dimensional skeleton of the person, and identifying the action for the product by detecting a movement of the posture of parts.
3. The non-transitory computer-readable recording medium according to claim 1 , wherein the process further includes determining the degree of interest of the person in the product, on a basis of the identified action.
4. The non-transitory computer-readable recording medium according to claim 3 , wherein the determining includes
performing determination by use of whether the product was purchased, a number of times of touching the product, or a period of time of staying in front of the product.
5. The non-transitory computer-readable recording medium according to claim 1 , wherein the detection rule specifies the action performed by the person for every combination of the size of the product and the degree of interest.
6. The non-transitory computer-readable recording medium according to claim 1 , wherein the process further includes
counting a number of persons who purchased the product or did not purchase the product, among customer service target persons each identified with the action, as a result of a customer service performed using the detection rule, and
updating the degree of interest correlated with the action in the detection rule, on a basis of a counted result.
7. A generation method comprising executed by a computer, the generation method:
obtaining each piece of image data captured by a camera installed in a store within a period of time from entering until exiting of a person at a store;
generating joint positions of a three-dimensional skeleton of the person by inputting each piece of the image data into a trained machine learning model;
identifying joint positions of a skeleton related to the person by analyzing the each piece of the image data;
identifying the product on the basis of the each piece of the image data and coordinates of the product in the each piece of the image data and a size of identified product by referring to a product database storing information about products;
determining a degree of interest of the person in the product, on a basis of the identified action for every size of the product;
identifying, as an action which indicates the degree of interest of the person in the product, the action performed by the person to the product in the store from the entering until the exiting for every size of the product, on a basis of the joint positions of the skeleton;
and generating a detection rule that correlates the identified action and size of the product and the determined degree of interest.
8. An information processing device comprising:
a memory; and
a processor coupled to the memory and configured to:
obtain each piece of image data captured by a camera installed in a store within a period of time from entering until exiting of a person at a store;
generate joint positions of a three-dimensional skeleton of the person by inputting each piece of the image data into a trained machine learning model;
identify joint positions of a skeleton related to the person by analyzing the each piece of the image data;
identify the product on the basis of the each piece of the image data and coordinates of the product in the each piece of the image data and a size of identified product by referring to a product database storing information about products;
determine a degree of interest of the person in the product, on a basis of the identified action for every size of the product;
identify, as an action which indicates the degree of interest of the person in the product, the action performed by the person to the product in the store from the entering until the exiting for every size of the product, on a basis of the joint positions of the skeleton; and
generate a detection rule that correlates the identified action and size of the product and the determined degree of interest.
9. The information processing device according to claim 8 , wherein the processor configured to judge a posture of parts in the person based on the joint positions of the three-dimensional skeleton of the person, and identifying the action for the product by detecting a movement of the posture of parts.
10. The information processing device according to claim 8 , wherein the processor configured to determine the degree of interest of the person in the product, on a basis of the identified action.
11. The information processing device according to claim 10 , wherein the processor configured to perform determination by use of whether the product was purchased, a number of times of touching the product, or a period of time of staying in front of the product.
12. The information processing device according to claim 8 , wherein the detection rule specifies the action performed by the person for every combination of the size of the product and the degree of interest.
13. The information processing device according to claim 8 , wherein the processor configured to:
count a number of persons who purchased the product or did not purchase the product, among customer service target persons each identified with the action, as a result of a customer service performed using the detection rule, and
update the degree of interest correlated with the action in the detection rule, on a basis of a counted result.