Trajectory Analysis for Shopper Intention Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video analytic technologies are inadequate in recognizing fine details of human behavior in retail spaces, relying heavily on resource-intensive data collection and requiring additional visual cues, which limits the ability to accurately infer shopper intentions from trajectory data.
Innovation Solution
A method and system that utilizes trajectory data from video sequences to interpret shopper behavior by extracting dynamic features such as position, speed, and orientation, employing rule-based and probabilistic graphical models like Hidden Markov Models to infer shopper intentions without the need for additional visual cues or dedicated communication devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If current video analytic technology is used to recognize fine detail of human behavior, then measurement precision is improved, but device complexity and resource requirements increase significantly
Solution Approach 1:
The patent extracts only the trajectory data from the full video sequence, filtering out unnecessary visual details and focusing solely on motion paths. This extraction approach enables accurate shopper behavior recognition without requiring complex video analysis systems, resolving the contradiction between measurement precision and device complexity.
Solution Approach 2:
The patent creates a simplified representation of shopper behavior using trajectory data as a copy of the essential movement information. This copying approach preserves the key behavioral patterns while eliminating the need for complex visual processing, achieving accurate recognition with reduced system complexity.
2Measurement precision
If additional visual cues and dedicated communication devices are used to infer shopper intentions, then measurement precision is improved, but device complexity and cost increase
Solution Approach 1:
The patent makes trajectory data serve multiple functions: it simultaneously provides spatial location, movement direction, temporal patterns, and behavioral context. This multi-functionality eliminates the need for separate visual cues and communication devices, achieving accurate intention inference with a single data type while reducing overall system complexity.
Solution Approach 2:
The trajectory data itself contains sufficient information to infer shopper intentions without requiring external visual cues or additional sensors. The data serves itself by encoding all necessary behavioral information directly in the motion paths, eliminating dependency on other device types and reducing system complexity.
3Device complexity
If trajectory data alone is used to analyze shopper behavior, then device complexity is reduced, but measurement precision of fine behavior details deteriorates
Solution Approach 1:
The patent applies dynamic analysis to trajectory data, examining changes in motion patterns over time such as speed variations, direction changes, and stopping behaviors. This dynamic approach enables detection of fine behavioral details like product interest, comparison actions, and purchase intentions, achieving high measurement precision while maintaining system simplicity.
Solution Approach 2:
The patent transforms static trajectory positions into dynamic behavioral insights by analyzing the temporal and spatial relationships between consecutive positions. This dimensional transformation from simple coordinates to behavior patterns enables precise detection of fine details while keeping the system simple, as it processes only the trajectory data without additional sensors.
Data Source
AI summary
The present invention is a method and system for automatically recognizing which products a shopper intends to find or purchase based on the shopper's trajectory in a retail aisle. First, the system detects and tracks the person to generate the trajectory of the shopper. Then some of the dynamic features are extracted from the shopper trajectory. The shopper trajectory features of a given trajectory are typically the positions, the motion orientations, and speeds at each point of the trajectory. A shopper behavior model is designed based on some of the primitive actions of shoppers. The last step of the method is to analyze a given shopper trajectory to estimate the shopper's intention. The step either utilizes decision rules based on the extracted shopper trajectory features, or utilizes a trained Hidden Markov Model, to estimate the progression of the primitive actions from the trajectory. The decoded progression of the shopper behavior states is then interpreted to finally determine the shopper's intention.