Brand Visibility Quantification via Deep Neural Network Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automated methods fail to effectively quantify brand visibility and compliance in retail environments, as they lack the ability to extract detailed information from images or videos of brand displays, leading to inaccurate results and inefficiencies in marketing strategies.
Innovation Solution
A processor-implemented method using a deep neural networking model to recognize assets, determine brand associations, and compute visibility and compliance metrics by analyzing media content, including images, videos, or 3D models of retail environments, generating heatmaps and attention sequences to assess brand placement, size, color contrast, and distinctness.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual auditing is performed by human auditors to check display compliance, then detailed inspection can be conducted, but the process becomes very difficult and cumbersome and can only be performed on a sample basis
Solution Approach 1:
The patent replaces manual human auditing with an automated computer vision system that uses machine learning models to detect and analyze brand displays. The system automatically processes images and videos to identify assets, detect brands, and evaluate compliance metrics without human intervention, thereby eliminating the difficulties of manual auditing while maintaining high precision through algorithmic analysis.
2Productivity
If existing automated methods are used to quantify brand visibility, then some visibility metrics can be obtained, but they fail to extract fine level details such as position of brand and logo, color, and size of text
Solution Approach 1:
The patent segments the brand display analysis into multiple distinct components: asset detection, brand identification, logo positioning, text extraction, color analysis, and size measurement. Each component is handled by specialized processing modules that extract specific fine-level details independently, enabling comprehensive precision across all measurement dimensions while maintaining automated efficiency.
Solution Approach 2:
The system changes multiple parameters simultaneously to achieve detailed extraction: it detects spatial coordinates for position, analyzes RGB values for color, measures pixel dimensions for size, and identifies text content through OCR. By monitoring and processing multiple parameters in parallel, the system achieves fine-level detail extraction across all critical attributes of brand displays.
3Loss of time
If sample-based manual auditing is performed to check compliance, then some compliance data can be collected, but the results are projected based on sample which may result in inaccurate results
Solution Approach 1:
The patent implements continuous automated monitoring that processes all display instances without sampling. The system continuously captures and analyzes images and videos of brand displays across the retail environment, ensuring that every display is evaluated rather than relying on statistical projections from samples. This continuous action eliminates sampling errors while maintaining efficient automated operation.
Data Source
AI summary
A system for recognizing a plurality of assets in an environment, determining a brand associated with each of the plurality of assets using a deep neural networking model, and computing a brand visibility and a compliance metric for the brand is provided, The system (i) determines a location of a plurality of assets and type of each of the plurality of assets within the media content, (ii) determines a brand and at least one object from the brand associated with each of the plurality of assets, (iii) determines at least one attribute of the at least one determined object associated with the brand, (v) implements at least one compliance rule to the at least one attribute of the at least one object, (vi) automatically determines a brand visibility and a compliance metric for the brand within the environment based on attention sequence and heatmap corresponding to the media content.