Pose-Matched Fitting Room Video Comparison for Apparel Try-Ons

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current retail fitting room systems require shoppers to repeatedly try on multiple clothing or jewelry items to compare their appearance, which can be time-consuming and inefficient, as they need to hold up different items against a mirror to assess how they look, lacking a seamless method to compare multiple pieces simultaneously.

Innovation Solution

A system that records and matches video sequences of a shopper wearing different items, allowing for simultaneous comparison by creating a third sequence that remixes frames from the first sequence to match the second sequence, enabling near real-time playback of the two sequences side by side, with user interface controls for easy navigation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If shoppers repeatedly try on multiple clothing items to compare their appearance, then they can assess how each item looks, but the process becomes time-consuming and inefficient

Engineering Contradiction:
Improvevisual comparison accuracyVSAvoidshopping time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system pre-records video sequences of the shopper wearing different clothing items before the comparison session. By capturing all necessary visual data in advance, the system eliminates the need for repeated real-time try-ons during the comparison process, thereby reducing shopping time while maintaining visual comparison accuracy.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system creates digital copies (video sequences) of the shopper wearing different items and stores them for later comparison. Instead of requiring the shopper to physically try on each item repeatedly, the system uses pre-recorded video copies to enable seamless visual comparison, significantly reducing the time required while preserving assessment accuracy.

Inventive Principle:
Principle #26Copying

2Adaptability or versatility

If shoppers hold up different items against a mirror to assess appearance, then they can compare items, but they cannot simultaneously compare multiple pieces

Engineering Contradiction:
Improvecomparison capabilityVSAvoidoperational simplicity
Core Design Contradiction:
Adaptability or versatilityVSEase of operation

Solution Approach 1:

The system transitions from single-item sequential comparison (one dimension) to multi-item simultaneous comparison by displaying multiple video sequences side-by-side on a display screen. This dimensional expansion allows shoppers to view and compare multiple clothing items at once, enhancing comparison capability while maintaining ease of operation through intuitive visual presentation.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The system enables a single comparison interface to handle multiple clothing items simultaneously, making the comparison tool universal. Instead of requiring separate mirror-viewing actions for each item, the display system presents multiple items in a unified view, allowing shoppers to compare any number of items through a single operational interface.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Measurement precision

If the system records video sequences of the shopper wearing different items, then simultaneous comparison becomes possible, but the sequences must be matched for body and head poses to ensure accurate comparison

Engineering Contradiction:
Improvepose matching accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system replaces manual pose-matching operations with automated computer vision algorithms. Instead of requiring manual alignment or complex mechanical adjustment mechanisms, the system uses image processing and pose estimation algorithms to automatically match body and head poses between different video sequences, achieving high pose matching accuracy while avoiding excessive system complexity.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS8311339B2Method to robustly match images with similar body and head pose from image sequence
Publication Date: 2012.11.13 REGIONAL RESOURCES LTD
  • US8311339B2 patent drawing
  • US8311339B2 patent drawing
  • US8311339B2 patent drawing

AI summary

During shopping, a shopper looks at herself in a mirror to evaluate clothing, jewelry, etc. because the mirror can provide a third-person view of the item. One thing a mirror cannot do is to show how two different items look at the same time, because only one item can be tried on at a time. Because of this, shoppers will often repeatedly try on items and must compare the look from memory. To enable self-comparison visually rather than from memory, embodiments can detect matches between images from two separate recorded single camera video sequences (corresponding to two different fittings). The matched images can then be played back to the user in close visual proximity for easy comparison shopping.