Facial Portrait Segmentation for Collaborative Picture Editing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing image processing methods fail to ensure target user satisfaction due to varying interpretations of image processing, particularly in group photos where individuals focus on their own parts, leading to inefficient and time-consuming processing.

Innovation Solution

Perform face recognition on a target image, determine portrait areas corresponding to individual faces, and distribute these areas to different target users for online editing, ensuring each user's satisfaction with the processed image.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional image processing methods are used, then processing can be completed, but user satisfaction is low due to varying interpretations of processing requirements

Engineering Contradiction:
Improveuser satisfactionVSAvoidprocessing efficiency
Core Design Contradiction:
ReliabilityVSEase of operation

Solution Approach 1:

The patent divides the image processing task into separate portrait areas corresponding to different faces, and distributes these segments to different users for independent editing. This segmentation allows each user to focus on specific regions according to their preferences, improving both satisfaction and efficiency simultaneously.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies different processing operations to different portrait areas based on individual user requirements. Each user can perform specialized editing on their assigned portrait areas, allowing localized optimization of processing quality rather than applying uniform processing to the entire image.

Inventive Principle:
Principle #3Local quality

2Productivity

If multiple users process different parts of the image simultaneously, then processing efficiency improves, but coordination complexity increases

Engineering Contradiction:
Improveprocessing efficiencyVSAvoidsystem coordination
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The system segments the image into distinct portrait areas and assigns them to different users, enabling parallel processing. This segmentation strategy allows multiple users to work simultaneously on different portions of the image without interference, significantly improving productivity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces a server as an intermediary component that manages the distribution of portrait areas to users and coordinates the synthesis of edited results. This intermediary simplifies the coordination complexity by centralizing the management of multiple users and their respective editing tasks.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Speed

If face recognition and portrait area determination are performed automatically, then processing speed increases, but accuracy may decrease due to statistical estimation

Engineering Contradiction:
Improveprocessing speedVSAvoidportrait area accuracy
Core Design Contradiction:
SpeedVSMeasurement precision

Solution Approach 1:

The patent performs face recognition and portrait area determination as preliminary automated actions before user editing begins. This preliminary processing speeds up the overall workflow by pre-dividing the image into manageable portrait areas, allowing users to focus directly on editing rather than manual region selection.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system uses feedback from face recognition results to automatically determine portrait areas, and this determination can be adjusted based on user input during the editing process. The feedback mechanism allows the system to adapt to user needs while maintaining the speed benefits of automated initial processing.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP4195096B1Picture processing method and apparatus, device, and storage medium
Publication Date: 2025.10.01 DOUYIN VISION CO LTD
  • EP4195096B1 patent drawingFigure 1~2
  • EP4195096B1 patent drawingFigure 3~4

AI summary

The present disclosure provides a picture processing method and apparatus, a device, and a storage medium. Said method comprises: in response to a processing triggering operation for a target picture, performing facial recognition on the target picture, and determining a first human image area corresponding to a first face on the target picture; and then sending the first human image area to a first target user, and the first target user performing online editing on the first human image area online to obtain a first edited human image area. Finally, the first edited human image area is composited into the target picture, so as to complete the processing of the target picture. As the first human image area on the target picture is edited online by the corresponding first target user, the processed target picture can ensure the satisfaction degree of the first target user.