Nec corporation (20240119711). LEARNING APPARATUS, ESTIMATION APPARATUS, LEARNING METHOD, ESTIMATION METHOD, AND PROGRAM AND NON-TRANSITORY STORAGE MEDIUM simplified abstract

LEARNING APPARATUS, ESTIMATION APPARATUS, LEARNING METHOD, ESTIMATION METHOD, AND PROGRAM AND NON-TRANSITORY STORAGE MEDIUM

Organization Name

nec corporation

Inventor(s)

Hiroo Ikeda of Tokyo (JP)

LEARNING APPARATUS, ESTIMATION APPARATUS, LEARNING METHOD, ESTIMATION METHOD, AND PROGRAM AND NON-TRANSITORY STORAGE MEDIUM - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240119711 titled 'LEARNING APPARATUS, ESTIMATION APPARATUS, LEARNING METHOD, ESTIMATION METHOD, AND PROGRAM AND NON-TRANSITORY STORAGE MEDIUM

Simplified Explanation

The present invention is a learning apparatus that acquires learning data associating a training image with correct answer labels indicating the position of each person and the visibility of keypoints on their body. The learning unit then uses this data to learn an estimation model that can estimate the position of each person, visibility of keypoints, and related information for computing positions within a processing image.

• Acquisition unit acquires learning data associating a training image with correct answer labels
• Learning unit learns an estimation model based on the learning data
• Estimation model estimates information indicating the position of each person, visibility of keypoints, and related information for computing positions within a processing image

Potential Applications

This technology can be applied in various fields such as:

• Computer vision
• Image processing
• Human pose estimation

Problems Solved

This technology helps in:

• Improving accuracy in estimating the position of people in images
• Enhancing visibility detection of keypoints on the body

Benefits

The benefits of this technology include:

• Enhanced performance in image analysis tasks
• Increased efficiency in processing images with multiple people
• Improved understanding of human poses in images

Potential Commercial Applications

With its capabilities, this technology can be utilized in:

• Surveillance systems
• Sports analytics
• Healthcare imaging technologies

Possible Prior Art

One possible prior art for this technology could be the use of deep learning models for human pose estimation in images.

How does this technology compare to existing human pose estimation methods?

This article does not provide a direct comparison with existing human pose estimation methods.

What are the limitations of this technology in real-world applications?

This article does not address the potential limitations of this technology in real-world applications.

Original Abstract Submitted

the present invention provides a learning apparatus () including an acquisition unit () that acquires learning data associating a training image including a person with a correct answer label indicating a position of each person, a correct answer label indicating whether each of a plurality of keypoints of a body of each of the persons is visible in the training image, and a correct answer label indicating a position, within the training image, of the keypoint being visible in the training image among a plurality of the keypoints, and a learning unit () that learns, based on the learning data, an estimation model that estimates information indicating a position of each person, information indicating whether each of a plurality of the keypoints of each person included in a processing image is visible in the processing image, and information being related to a position of each of keypoints for computing a position, within the processing image, of the keypoint being visible in the processing image.