HumanSensing

PersonDetection

The person detection software can detect the position of individuals in an image. It can also specialize in detecting specific parts such as the upper body, head, or any other designated area.

Human Detection Examples

Examples of Use

Store Visitor Analysis

  • Counting the number of visitors
  • Measuring visitor flow
  • Measuring congestion levels

Factory Worker Analysis

  • Measuring movement paths

Intrusion Detection

  • Detecting entry into prohibited areas
Human Detection Main Function

Main Function

  • Can detect up to 50 people simultaneously and has the ability to track detected individuals.
  • Detection targets can be selected from the head, upper body, or whole body
Human Detection Features

Advantages

Through unique feature extraction and matching methods, we achieve high-precision and fast detection.

Human Detection Offerings

Offering Type

  • We provide the recognition library (SDK), sample programs for operation verification, and technical documents such as API specifications.
  • Supported platforms by default are Windows, PC-Linux, and ARM-Linux, but we can also accommodate other platforms based on specific needs.

SkeletonDetection

The skeletal detection software can detect the skeletons of individuals in an image. We also has the technology to estimate 3D skeletal information from regular images. Although this increases computational load, utilizing 3D information can enhance the accuracy of behavior estimation and other analyses.

Skeleton Detection Examples

Examples of Use

  • Driver Physique Detection
  • Driver Behavior Detection (in combination with object detection technology, estimates behaviors such as smoking or eating while driving)
  • Behavior Detection in Care Facilities (detects falls, bed falls, etc.)
  • Behavior Risk Assessment in Care Facilities (monitors changes in daily skeletal movements that may lead to risks)
  • Work Analysis of Factory Workers
  • Form Analysis in Sports
Skeleton Detection Main Function

Main Features

Obtains skeletal points of a person from an image with a specified position.

3D Pose Estimation

  • From camera footage (2D information), obtains 3D coordinate information, including depth, for joint points such as hands and feet.

Output Information of Estimation Results

  • Number of joint points: 17
  • Joint angles, etc.: 4 locations (elbows, knees, shoulders, waist)
Skeleton Detection Features

Advantages

  • 3D pose estimation can be achieved with a single camera without the need for multiple cameras, reducing capital investment costs.
  • Utilization of 3D pose estimation technology becomes possible in areas where the installation of multiple cameras is difficult or impractical
Skeleton Detection Offerings

Offering Type

  • We provide evaluation samples for Windows.
  • For libraries to be embedded in products, please contact us individually.
product