top of page

LIPSense™ 3D Body Pose Middleware

The True Body Motion Detection

資產 66_4x.png

The LIPSense™ 3D Body Pose SDK is an industry leading Real-time Full-body skeletal tracking with 18 specific joints middleware. Its tracking can include up to 10 people, with 360° motion tracking  for side view, back view, sitting and lying poses.​

Features That Matter Most

資產 50_4x.png

18

Joint full-body skeletal tracking

資產 52_4x.png

10

People
tracking

資產 51_4x.png

Swing & staggering Motion

資產 48_4x.png
資產 47_4x.png

Leading API Wrappers for Integration and Acceleration

Extended Full Featured Trial Time

Enjoy fully the complete featured SDK in Trial Mode !!!

  • Full-body Skeletal Tracking

  • Outputs RGB, Depth and Skeletal Image

  • Joint-tracking capability of 18 joints

  • Eyes * 2

  • Nose

  • Ears * 2

  • Neck

  • Shoulder * 2

  • Elbow * 2

  • Wrist * 2

  • Hip * 2

  • Knee * 2

  • Ankle * 2

  • People Tracking up to 10

  • Capture of Swinging and Staggering Motions

  • Support for multiple 3D camera mechanical positions

  • 90 Clockwise

  • 90 Counterclockwise

  • 180 Flat

  • Capability to detect cameras connected during runtime

  • Supporting both Industrial and Consumer 3D Depth Cameras

  • LIPSedge™ AE400 Series

  • LIPSedge™ DL Series

  • Intel® RealSense™ D400 Series

資產 49_4x.png

Complete Documentation

ROS ウェブサイト

Refer to our User Guides to know how easy it is to use

Read Documentation...

ROS ウェブサイト

Most recent updates, new features and implementations

Know more...

ROS ウェブサイト

Codes in our User Guide to show how friendly our SDK is

Test and Verify...

圖片8.png

Unity API Wrapper

Save development time and cost with LIPScan™ 3D Body Pose SDK

“ Extend the power of the world’s leading platform for creating and operating interactive, real-time 3D content.”

​Empowered with TensorRT™

LIPSense™ 3D Body Pose SDK  utilizes NVIDIA’s TensorRT enables you to optimize inference using techniques such as quantization, layer and tensor fusion, kernel tuning, and others on NVIDIA GPUs

資產 60_4x.webp

Speed up inference by 36x

NVIDIA TensorRT-based applications perform up to 36X faster than CPU-only platforms during inference

資產 61_4x.webp

Optimize inference performance

TensorRT, built on the NVIDIA CUDA® parallel programming model

資產 62_4x.webp

Accelerate every workload

TensorRT provides INT8 using quantization-aware training and post-training quantization and Floating Point 16 (FP16) optimizations for deployment of deep learning inference applications

資產 59_4x.webp

Deploy, run, and scale with Triton

TensorRT-optimized models can be deployed, run, and scaled with NVIDIA Triton™, an open-source inference serving software that includes TensorRT as one of its backends.

​Empowered with OpenVINO™

LIPSense™ 3D Body Pose SDK utilizes the Intel® OpenVINO™ toolkit, and when equipped with Intel® HD graphics, it allows the acceleration of the inference engine and enhance the visualization experience

圖片1.png
圖片2.png

Specifications and Requirements

Applications

  • Interactive exhibition or entertainment

        - Audience interaction detection

        - Body motion range, crowd position & pose detection

 

  • Workspace monitoring

       - Operator monitoring for safety assistant

       - Human machine collaboration

       - Work elements record and working time measurement

  • Athletes Motion Analysis

       - Operator ergonomics analysis

       - Human factor engineering

 

 

  • Therapists & fitness analysis

       - Posture analysis, evaluation, correction and tracking

The Middleware

LIPSense™ 3D Body Pose Demo

LIPSense™ 3D Body Pose Comparison

Successful Use Case

Workspace Monitoring

​

Requirement

  • Real-Time Human-Robot Collaboration

  • Collision –free Path Planning for Mobile Robot

​​

Challenge​

  • Human Behavior Monitor & Prediction

  • Blind Spot for AMR onboard sensor

​

Benefits of LIPSense™ 3D Body Pose SDK​

  • Real-time & High-Performance 3D Skeleton Tracking

  • Enabling Workspace – Human Tracking Overview for AMR coordination.

3D Virtual Fence for Factory Automation
​
LIPSense™ Virtual Fence is a 3D Sensing Solution for Workplace Safety.
​
The cost of workplace accidents and operation hazards in automated factories can generate high costs when left unattended. A proper solution implementing the right tools is a key factor to reduce costs and improve efficiency.
​
Set in an automated factory line, LIPSense™ Virtual Fence can be integrated into  existing factory robotic systems, detects if on-site factory workers’ movements and presence fall within areas that could endanger them.

資產 64_4x.png

今すぐ入手

You can now buy a sample of LIPSedge AE Series today with free world-wide shipping and start migrating your RealSense application for industrial deployment!

bottom of page