Method and system for measuring packaging effectiveness using video-based analysis of in-store shopper response
View Patent ↗The present invention is a method and system for measuring packaging effectiveness based on video-based analysis of people in the vicinity of a plurality of package concepts. The present invention measures the response of people to the different packaging concepts, wherein the quantified response measurement shows the ability of the packaging concepts to attract, engage, and motivate people to purchase. The video-based analysis comprises behavior analysis, demographic analysis, and emotion change detection in response to the package effectiveness. The present invention captures a plurality of input images of the people in the vicinity of a plurality of package concepts by a plurality of means for capturing images. The plurality of input images is processed in order to understand the effectiveness of packaging design and changes. The processes are based on a novel usage of a plurality of computer vision technologies to analyze the visual characterization of the people from the plurality of input images in relation to the packaging design and changes.
1. A method for measuring packaging effectiveness based on video-based analysis of a plurality of persons using a control and processing system that executes computer vision algorithms on a plurality of input images transferred to the control and processing system through a means for video interface, comprising the following steps of:
a) processing a behavior analysis of a person based on a path analysis of the person by executing a computer vision-based tracking algorithm on the plurality of input images to obtain information about the person's behavior with package concepts using the control and processing system,
b) processing a segmentation of the person, including demographic analysis, by applying a computer vision-based demographic classification algorithm to the plurality of input images in relation to the package concepts,
c) detecting an emotion change of the person,
d) producing indices along various interaction stages of the person with each package concept based on the behavior analysis, the segmentation, and the emotion change detection,
e) sending the indices to a module for an association of measurements and analyses, which performs a package effectiveness measurement, including a comparison of the indices among a plurality of package concepts, and
f) producing comparison results of the indices in relation to the segmentation,
g) organizing the measurement of the packaging effectiveness based on the segmentation and trips of the plurality of persons,
wherein the segmentation comprises demographic segments, including age-range, gender and ethnicity,
wherein the trips comprise a shopping pattern of said plurality of persons, and wherein the indices comprise reach to attract index, attract to engage index, and engage to motivate index.
2. The method according to claim 1 , wherein the method further comprises a step of measuring packaging effectiveness in a retail context by documenting shopper interactions with the package concepts, based on the behavior analysis and the segmentation,
wherein the retail context comprises information from measuring several package concepts in a physical location along with other package concepts from competitors while taking into consideration temporal information, including time of year, and other in-store factors, like promotion of a product, and
wherein the retail context further comprises a physical location of the package concepts, local influence, trends, and a prior knowledge about the package concepts.
3. The method according to claim 1 , wherein the method further comprises a step of comparing the measurements among the plurality of package concepts synchronously in a plurality of store locations,
wherein the plurality of store locations have similar store layout, shopper demographic composition, shopper psychographics, and product location.
4. The method according to claim 1 , wherein the method further comprises a step of comparing the measurements among the plurality of package concepts asynchronously in a store and a product location during different time periods,
wherein the plurality of package concepts differ at most on one factor during the different time periods.
5. The method according to claim 1 , wherein the method further comprises steps of:
a) changing a feature in the package concepts, and
b) measuring the effectiveness of the newly changed package concepts,
wherein the package concepts are compared with each other along several measures, based on package effectiveness decision criteria, to determine the effectiveness, and
wherein the package effectiveness decision criteria comprise comparison of conversion ratios among indices along each stage of a shopper interaction funnel, effectiveness count per shopper segment, prominent value changes induced by the newly changed feature in the package concepts.
6. An apparatus for measuring packaging effectiveness based on video-based analysis of a plurality of persons, comprising:
a) at least a means for capturing images in the vicinity of a plurality of package concepts, wherein the means for capturing images captures a plurality of input images of said plurality of persons,
b) at least a means for video interface that is connected to the means for capturing images, and
c) a control and processing system that executes computer vision algorithms on the plurality of input images for tracking at least a person,
performing a path analysis of the person based on the tracking,
processing a behavior analysis of the person based on the path analysis to obtain information about the person's behavior, including engagement behavior with the package concepts,
processing a segmentation of the person, including demographic analysis in relation to the package concepts,
detecting an emotion change of the person,
producing indices along various interaction stages of the person with each package concept based on the behavior analysis, the segmentation, and the emotion change detection,
sending the indices to a module for an association of measurements and analyses, which performs a package effectiveness measurement, including a comparison of the indices among the plurality of package concepts, and
producing comparison results of the indices in relation to the segmentation,
organizing the measurement of the packaging effectiveness based on the segmentation and trips of the plurality of persons,
wherein the segmentation comprises demographic segments, including age-range, gender and ethnicity,
wherein the trips comprise a shopping pattern of said plurality of persons, and wherein the indices comprise reach to attract index, attract to engage index, and engage to motivate index.
7. The apparatus according to claim 6 , wherein the apparatus further comprises means for measuring packaging effectiveness in a retail context by documenting shopper interactions with the package concepts, based on the behavior analysis and the segmentation,
wherein the retail context comprises information from measuring several package concepts in a physical location along with other package concepts from competitors while taking into consideration temporal information, including time of year, and other in-store factors, like promotion of a product, and
wherein the retail context further comprises a physical location of the package concepts, local influence, trends, and a prior knowledge about the package concepts.
8. The apparatus according to claim 6 , wherein the apparatus further comprises means for comparing the measurements among the plurality of package concepts synchronously in a plurality of store locations,
wherein the plurality of store locations have similar store layout, shopper demographic composition, shopper psychographics, and product location.
9. The apparatus according to claim 6 , wherein the apparatus further comprises means for comparing the measurements among the plurality of package concepts asynchronously in a store and a product location during different time periods,
wherein the plurality of package concepts differ at most on one factor during the different time periods.
10. The apparatus according to claim 6 , wherein the apparatus further comprises:
a) means for changing a feature in the package concepts, and
b) means for measuring the effectiveness of the newly changed package concepts,
wherein the package concepts are compared with each other along several measures, based on package effectiveness decision criteria, to determine the effectiveness, and
wherein the package effectiveness decision criteria comprise comparison of conversion ratios among indices along each stage of a shopper interaction funnel, effectiveness count per shopper segment, prominent value changes induced by the newly changed feature in the package concepts.