On-Screen Control Image Selection for Make-Up Simulation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional make-up simulation systems face challenges in accurate detection of characteristic coordinates and latency issues when users are not directly facing the camera, leading to low-quality make-up simulation results due to the need for on-screen control and real-time processing constraints.

Innovation Solution

The system captures video frames and selects a suitable still image for make-up control by detecting characteristic coordinates, ensuring the user's face is directly facing the camera, reducing latency by using a still image for control mode operations, allowing for precise cosmetic application and improved user experience.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If real-time video processing is performed for make-up simulation, then the user can see make-up effects in real-time, but latency occurs due to computation requirements

Engineering Contradiction:
Improvereal-time processing speedVSAvoidprocessing latency
Core Design Contradiction:
SpeedVSLoss of time

Solution Approach 1:

The system pre-processes and stores multiple candidate still images with pre-detected characteristic coordinates before the user needs to perform make-up control. When the user enters control mode, the system immediately selects a suitable pre-processed image without performing detection at that moment, thereby eliminating detection latency while maintaining real-time visual feedback capability.

Inventive Principle:
Principle #10Preliminary action

2Ease of operation

If the user operates the control panel while facing the monitor, then make-up control can be performed, but the user's face does not directly face the camera resulting in lower detection accuracy

Engineering Contradiction:
Improvecontrol operation convenienceVSAvoidcharacteristic coordinate detection accuracy
Core Design Contradiction:
Ease of operationVSMeasurement precision

Solution Approach 1:

The system captures multiple still images during a preparation phase when the user is facing the camera, pre-detects characteristic coordinates for each image, and stores them as candidates. This allows the system to have high-accuracy pre-detected coordinates ready before the user switches to control mode, resolving the conflict between operation convenience and detection accuracy.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system creates a copy of the user's face image captured when facing the camera, and performs make-up simulation on this copied image during control operations. This allows the user to control make-up while facing the monitor, using the pre-captured high-quality face image as the base, thereby maintaining both detection accuracy and operational convenience.

Inventive Principle:
Principle #26Copying

3Reliability

If characteristic coordinates are detected from video frames during control mode, then the latest face position can be used, but detection accuracy decreases when the user is not facing the camera

Engineering Contradiction:
Improveface position accuracyVSAvoidcontrol mode usability
Core Design Contradiction:
ReliabilityVSEase of operation

Solution Approach 1:

The system performs characteristic coordinate detection in advance during a face-detection phase, storing multiple candidate images with their detected coordinates. When control mode is activated, the system selects from these pre-detected candidates rather than performing new detection, ensuring both high accuracy and smooth control operation without real-time detection delays.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP2988486B1Selection of a suitable image for receiving on-screen control
Publication Date: 2020.03.11 PANASONIC INTELLECTUAL PROPERTY MANAGEMENT CO LTD
  • EP2988486B1 patent drawingFigure 1
  • EP2988486B1 patent drawingFigure 2
  • EP2988486B1 patent drawingFigure 3

AI summary

Method possesses: operation mode checking step (S102) of selecting one of update mode and hold mode; when in update mode, processing-target image updating step (S108) of updating video as update processing-target image; when in hold mode, characteristic coordinate position detecting step (S104) of detecting characteristic coordinates in input video; characteristic coordinate position examining step (S105) of examining positions of the characteristic coordinates; processing-target image selecting step (S106) of selecting processing-target image; and processing-target image holding step (S109) of holding current video frame as held processing-target image.