IP Library Granted Patent US 9,406,024
Granted Patent B1
US 9,406,024 · App. 14/618,503 · Granted Aug 2, 2016

System and method for color paint selection and acquisition

Inventors: Michael S. Gordon (Yorktown Heights, NY); James R. Kozloski (New Fairfield, CT); Peter K. Malkin (Yorktown Heights, NY); Clifford A. Pickover (Yorktown Heights, NY)
Assignee: International Business Machines Corporation
G06N5/04G06N99/005
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 9,406,024
App. No.
14/618,503
Granted
Aug 2, 2016
Kind
B1
Abstract

A system and method and computer program product for paint color recommendation. The system obtains measures of an environment to be painted and trains a learned model to input data received from customers including data representing each customer's initial color paint and pigment selection, and one or more of: a customer perceptual, a customer context, and environment measure (P/C/E data) to generate a sparse matrix. One or more paint vendors may then use the generated sparse matrix to determine a color pigment recommendation from a pigments color space for a customer. From a user selected color/pigment, and using the learned model, the system maps the selection, together with the user's P/C/E data back to the color/pigments space. User feedback representing a degree of satisfaction that the recommended color pigment applied to the user environment has matched the user's initial color paint and color pigment selection is elicited.

Claims (54)

1. A paint color recommendation system for multiple paint vendors, said system comprising:

a memory storage device storing a program of instructions;

a processor device receiving said program of instructions to configure said processor device to:

receive input data representing an initial paint color or pigment selection from a user;

receive further input data representing one or more of a user's perceptual/cognitive/environmental (P/C/E) context;

train, using a machine learning technique, a sparse regression model to received input data from multiple user's including each user's initial color paint and color pigment selection data, received user profile data, a received user context data, and a received measures data to generate a sparse matrix;

map, using the generated sparse matrix, said user's initial paint color or pigment selection and user context data to a pigments color space;

determine, based on said map, a color pigment or pigment mixture from said pigments color space for recommendation to the user; and

communicate data representing said color pigment or mixture recommendation to a device.

2. The paint color recommendation system of claim 1 , wherein said context may include the user's cognitive context obtained from social network information, said processor device being further configured to:

conduct a search in a social networking site or network infrastructure to obtain user context information comprising one or more of: the user's interest in art, design goods, music and reading material, and the user's purchase history of art and design goods, music and reading material.

3. The paint color recommendation system of claim 2 , wherein said color pigment recommendation data specifies a unique combination of and amounts of paint colors and color pigments, said paint vendor using said specified color pigment recommendation data to mix a test supply of color paint of the recommended color pigment for said user.

4. The paint color recommendation system of claim 3 , wherein said user applies said test supply of color paint of the recommended color pigment data to applies said supplied paint to said environment, said processor device is further configured to:

receive from said user a feedback data representing a degree of satisfaction that the recommended color pigment applied to the user environment has matched the user's initial color paint and color pigment selection; and

update said learned model with a positive or negative label based on said received user feedback data.

5. The paint color recommendation system of claim 4 , wherein to update said learned model, said processor device is further configured to:

apply, in the model, a positive label to an output recommended color pigment or pigment mixture based on a received favorable user feedback data, associated with the received input initial color paint and color pigment selection, and said one or more user P/C/E context data.

6. The paint color recommendation system of claim 1 , wherein said processor device is configured to receive user profile data representing one or more of: a gender, a socioeconomic status; an age of said user.

7. The paint color recommendation system of claim 1 , wherein said processor device is configured to receive user context data representing one or more of:

a picture or real-time image of a room or environment to be painted with said color paint and color pigment selection;

a result of an administered psychophysical assessment of the user's color perception to obtain components of the user's color perception; and

a measure associated with a feature of an environment in which the selected paint color is to be applied to a surface thereof.

8. The paint color recommendation system of claim 7 , wherein said measure associated with a feature of an environment to be painted comprises one or more of:

an average ambient light level as recorded over a period of time of a room or environment; and

an amount and color of items or furnishings to be located in said room or environment to be painted with a color paint and color pigment selection.

9. The paint color recommendation system of claim 1 , wherein to map said user's initial color paint and color pigment selection, said user context, and environment measure data to a pigments color space, said processor device is further configured to:

map one or more color components of the painted environment, a measure of the user's color perception, and components of the user's context to one or more of: a mixture of wavelengths corresponding to a desired color, a mixture of pigments corresponding to a desired wavelength of light sufficient to produce a desired color, and a base paint into which pigments are mixed.

10. A computer program product comprising a computer readable storage medium tangibly embodying a program of instructions executable by the computer for recommending paint color for multiple paint vendors, the program of instructions, when executing, performing the following steps:

receiving input data representing an initial paint color or pigment selection from a user;

receiving further input data representing one or more of a user's perceptual/cognitive/environmental (P/C/E) context, a user's environmental context associated with an environment in which the selected paint color is to be applied to a surface thereof;

training, using a machine learning technique, a sparse regression model to received input data from multiple user's including each user's initial color paint and color pigment selection data, received user profile, received user context data, and received measures data to generate a sparse matrix;

mapping, using the generated sparse matrix, said user's initial paint color or pigment selection and user context data to a pigments color space;

determining, based on said map, a color pigment or pigment mixture from said pigments color space for recommendation to the user; and

communicating data representing said color pigment or mixture recommendation to a device.

11. The computer program product of claim 10 , wherein said context may include the user's cognitive context obtained from social network information, said method further comprising:

conducting a search in a social networking site or network infrastructure to obtain user context information comprising one or more of: the user's interest in art, design goods, music and reading material, and the user's purchase history of art and design goods, music and reading material.

12. The computer program product of claim 11 , wherein said color pigment recommendation data specifies a unique combination of and amounts of paint colors and color pigments, said paint vendor using said specified color pigment recommendation data to mix a test supply of color paint of the recommended color pigment for said user, wherein said user applies said test supply of color paint of the recommended color pigment data to a surface within said environment, said method further comprising:

receiving from said user a feedback data representing a degree of satisfaction that the recommended color pigment from said test supply and applied to the user environment has matched the user's initial color paint and color pigment selection; and

updating said learned model with a positive or negative label based on said received user feedback data.

13. The computer program product of claim 12 , wherein said updating said learned model comprises:

applying, in the model, a positive label to an output recommended color pigment or pigment mixture based on a received favorable user feedback data, associated with the received input initial color paint and color pigment selection, and said one or more user P/C/E context data.

14. The computer program product of claim 10 , wherein said method further comprises:

receiving user profile data representing one or more of: a gender, a socioeconomic status; an age of said user.

15. The computer program product of claim 10 , wherein said method further comprises:

receiving user context data representing one or more of:

a picture or real-time image of a room or environment to be painted with said color paint and color pigment selection;

a result of an administered psychophysical assessment of the user's color perception to obtain components of the user's color perception; and

a measure associated with a feature of an environment in which the selected paint color is to be applied to a surface thereof.

16. The computer program product of claim 10 , wherein said method further comprises:

receiving environment measure data representing one or more of:

an ambient light level as recorded over a period of time of a room or environment to be painted with a color paint and color pigment selection; and

an amount and color of items to be located in said room or environment to be painted with a color paint and color pigment selection.

17. The computer program product of claim 10 , wherein said mapping of said user's initial color paint and color pigment selection, said user context, and environment measure data to a pigments color space further comprises:

mapping one or more color components of the painted environment, a measure of the user's color perception, and components of the user's context to one or more of: a mixture of wavelengths corresponding to a desired color, a mixture of pigments corresponding to a desired wavelength of light sufficient to produce the desired color, and a base paint into which pigments are mixed.

Assignments (7)
SECURITY AGREEMENT Recorded May 20, 2026
From: WAYFAIR LLC
To: U.S. BANK TRUST COMPANY, NATIONAL ASSOCIATION, AS NOTES COLLATERAL AGENT
Reel/Frame 075591/0399 →
SECURITY INTEREST Recorded Nov 10, 2025
From: WAYFAIR LLC
To: U.S. BANK TRUST COMPANY, NATIONAL ASSOCIATION, AS NOTES COLLATERAL AGENT
Reel/Frame 073514/0326 →
SECURITY AGREEMENT Recorded Mar 13, 2025
From: WAYFAIR LLC
To: U.S. BANK TRUST COMPANY, NATIONAL ASSOCIATION, AS NOTES COLLATERAL AGENT
Reel/Frame 070513/0542 →
SECURITY AGREEMENT Recorded Oct 10, 2024
From: WAYFAIR LLC
To: U.S. BANK TRUST COMPANY, NATIONAL ASSOCIATION, AS NOTES COLLATERAL AGENT
Reel/Frame 069143/0399 →
SECURITY AGREEMENT Recorded Mar 24, 2021
From: WAYFAIR LLC
To: CITIBANK, N.A.
Reel/Frame 055708/0832 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Oct 29, 2019
From: INTERNATIONAL BUSINESS MACHINES CORPORATION
To: WAYFAIR LLC
Reel/Frame 050867/0899 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 10, 2015
From: GORDON, MICHAEL S.; KOZLOSKI, JAMES R.; MALKIN, PETER K; PICKOVER, CLIFFORD A.
To: INTERNATIONAL BUSINESS MACHINES CORPORATION
Reel/Frame 034930/0650 →