Multi-Person Pose Scoring for Reference-Person Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing image analysis techniques fail to effectively handle images containing multiple individuals, lacking the ability to identify and highlight a reference person with a superior pose among them.

Innovation Solution

A pose analyzing apparatus and method that estimates poses, computes pose scores, and detects a reference person with a higher quality pose among multiple individuals in an image, utilizing a processor and memory to execute instructions for image acquisition, pose estimation, score computation, and reference detection.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If image analysis techniques are applied to images with multiple persons, then the ability to analyze class status is improved, but the capability to identify reference persons with superior poses is lost

Engineering Contradiction:

Improveability to handle multi-person imagesVSAvoidpose quality evaluation accuracy

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The system segments the image analysis process into distinct stages: first detecting all persons in the image, then individually analyzing pose characteristics for each detected person, and finally evaluating pose quality separately for each individual. This segmentation enables the system to handle multi-person images while maintaining accurate pose quality evaluation for each person.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system introduces an intermediary pose quality evaluation mechanism that compares each person's pose characteristics against reference pose data. This intermediary evaluation layer enables the identification of reference persons with superior poses by mediating between raw pose detection and final quality assessment.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Loss of information

If pose estimation is performed for each person in multi-person images, then comprehensive pose analysis is improved, but computational complexity increases

Engineering Contradiction:

Improvepose information completenessVSAvoidprocessing system complexity

Core Design Contradiction:

Loss of informationVSDevice complexity

Solution Approach 1:

The system performs preliminary person detection and pose characteristic extraction for all persons in the image before conducting detailed pose quality evaluation. By preparing pose information in advance for each detected person, the system reduces the computational burden of the subsequent quality evaluation stage while ensuring comprehensive pose information is captured.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system applies different processing depths to different persons based on their relevance. Reference persons with superior poses receive more detailed analysis while other persons undergo streamlined evaluation. This local quality approach maintains comprehensive pose information for all persons while reducing overall computational complexity through selective detailed analysis.

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS20250336177A1Pose analyzing apparatus, pose analyzing method, and non-transitory computer-readable storage medium

Publication Date: 2025.10.30 NEC CORP

AI summary

A pose analyzing apparatus acquires a target image and target person information. The target image includes two or more persons who do arbitrary thing, such as giving a performance, doing exercises, playing music instruments, etc. The pose analyzing apparatus estimates a pose for each person and computes, for each person, a pose score that represents quality of pose of the person. The pose analyzing apparatus detects one or more reference persons whose pose score is greater than the pose score of the target person, and outputs reference information that indicates the reference person.