Automatic image-based recommendations using a color palette
US-2015379733-A1 · Dec 31, 2015 · US
US2016267681A1 · US · A1
| Field | Value |
|---|---|
| Publication number | US-2016267681-A1 |
| Application number | US-201514745584-A |
| Country | US |
| Kind code | A1 |
| Filing date | Jun 22, 2015 |
| Priority date | Mar 11, 2015 |
| Publication date | Sep 15, 2016 |
| Grant date | — |
A practical reading order for non-experts. Skip the full description unless you need deep technical detail.
What the patent document calls the invention.
A short plain-language summary of the technical disclosure.
Who owns or filed the patent and who is credited as inventor.
Filing, priority, publication, and grant dates set the timeline.
The legal scope of protection — read this for what is actually claimed.
Technology tags used to group this patent with similar filings.
Prior art links and similar publications in this corpus.
Official abstract text for this publication.
A mechanism is provided for generating a set of color palettes for a product to be marketed. Color-related information of each image in a set of images is extracted. Using a user provided targeted product, targeted brand, and targeted brand message(s) to be marketed, computational logic is applied to generate a set of new color palettes for use in product design or product packaging design. The set of new color palettes is presented to the user. Responsive to the user selecting a color palette from the set of new color palettes, a color analysis of the selected color palette is performed. The results of the color analysis are then presented to the user.
Opening claim text (preview).
1 . A method, in a data processing system, for generating a set of color palettes for a product, the method comprising: extracting, by a processor in the data processing system, color-related information of each image in a set of images; using a user provided targeted product, targeted brand, and targeted brand message(s) to be marketed, applying, by the processor, computational logic to generate a set of new color palettes for use in product design or product packaging design; presenting, by the processor, the set of new color palettes to the user; responsive to the user selecting a color palette from the set of new color palettes, performing, by the processor, a color analysis of the selected color palette; and presenting, by the processor, results of the color analysis to the user. 2 . The method of claim 1 , wherein the set of images are images of different products and brands that are mined, by the processor, to learn about all colors and color combinations that frequently appear among the products and the brands. 3 . The method of claim 1 , wherein applying the computational logic to generate the set of new color palettes for use in the product design or the product packaging design further comprises: identifying, by the processor, a set of color categories (CC) that exist in product images associated with the targeted product; identifying, by the processor, a set of brand colors (BC) that exist in brand images associated with the targeted brand; identifying, by the processor, a set of inspirational colors (IC) of the targeted brand message(s) from a color-message graph; utilizing the set of images, identifying, by the processor, a set of universal colors (UC) that have co-appeared with at least one of the inspirational colors (IC); identifying, by the processor, an intersection of the set of brand colors (BC) and the set of universal colors (UC) thereby forming a set of junction colors (JC); and utilizing the set of color categories (CC) and the set of inspirational colors (IC) as the inspiration, taking optional color preferences from the user, and leveraging the set of junction colors (JC), outputting, by the processor, a final set of color combinations as the set of new color palettes. 4 . The method of claim 3 , wherein the color-message graph is constructed to represent messages conveyed by colors detected from the set of images, wherein the color-message graph captures an interrelationship among the messages, and wherein the interrelationship is at least one of an synonymic relationship or an antonymic relationship. 5 . The method of claim 3 , wherein the set of new color palettes are generated by randomly selecting a top color from the set of junction colors (JC) that fall into each of color category (CC) in the set of color categories (CC), and subsequently performing combinations and pruning. 6 . The method of claim 1 , wherein performing the color analysis of the selected color palette further comprises: identifying, by the processor, a colorfulness that measures the perceived intensity of colors contained in the selected color palette to human eyes. 7 . The method of claim 1 , wherein performing the color analysis of the selected color palette further comprises: identifying, by the processor, a color spreadness that measures how widely colors in the selected color palette are spread across a color wheel. 8 . The method of claim 1 , wherein performing the color analysis of the selected color palette further comprises: assessing, by the processor, a surprise factor for the selected color palette, wherein assessing the surprise factor is performed by calculating a cognitively-inspired Bayesian surprise. 9 . The method of claim 1 , wherein performing the color analysis of the selected color palette further comprises: identifying, by the processor, a set of synonymic messages to the targeted brand message(s) using a color-message graph. 10 . The method of claim 1 , wherein performing the color analysis of the selected color palette further comprises: identifying, by the processor, a set of antonymic messages to the targeted brand message(s) using a color-message graph. 11 - 20 . (canceled)
Drawing of charts or graphs · CPC title
Texturing; Colouring; Generation of textures or colours (retouching, inpainting or scratch removal G06T5/77) · CPC title
Color image · CPC title
Physics · mapped topic
Market modelling; Market analysis; Collecting market data · CPC title
Related publications grouped by family.
Answers are generated from the same data shown on this page.