Details

Project TitleOpenPose - Realtime Multiperson 2D Keypoint Detection from Video
Track CodeOpenPose
Short Description

OpenPose is a library for real-time multi-person keypoint detection and multi-threading written in C++ using OpenCV and Caffe*, authored by Gines Hidalgo, Zhe Cao, Tomas Simon, Shih-En Wei, Hanbyul Joo and Yaser Sheikh.

OpenPose represents the first real-time system to jointly detect human body, hand and facial keypoints (in total 130 keypoints) on single images. In addition, the system computational performance on body keypoint estimation is invariant to the number of detected people in the image.

Library main functionality:

  • Multi-person 15 or 18-keypoint body pose estimation and rendering. Running time invariant to number of people on the image.
  • Multi-person 2x21-keypoint hand estimation and rendering. Note: In this initial version, running time linearly depends on the number of people on the image.
  • Multi-person 70-keypoint face estimation and rendering.
  • Flexible and easy-to-configure multi-threading module.
  • Image, video, and webcam reader.
  • Able to save and load the results in various formats (JSON, XML, PNG, JPG, ...).
  • Small display and GUI for simple result visualization.
  • All the functionality is wrapped into a simple-to-use OpenPose Wrapper class.

The pose estimation work is based on the C++ code from the ECCV 2016 demo, "Realtime Multiperson Pose Estimation", Zhe Cao, Tomas Simon, Shih-En Wei, Yaser Sheikh.

* It uses Caffe, but the code is ready to be ported to other frameworks (Tensorflow, Torch, etc.).

AbstractNone
 
TagsOpenPose
 
Posted DateJul 29, 2017 7:00 AM

Contact

Name
David Graham

Results

Body Estimation

Body + Hands


Note: ***Current commercial software does not contain Face Estimation***

(Facial Estimation is expected to be re-integrated by December 2017)

Body + Face Estimation

Body + Hands + Face Estimation

Related Publications

Non-Exclusive Commercial License

Here is a copy of the license template: https://flintbox.com/file/download/10591

Important points from the license:

  • The non-exclusive commercial license requires a non-refundable $25,000 USD annual royalty.
  • The non-exclusive commercial license cannot be used in the field of Sports. ("Sports" shall mean any and all athletic competition between individuals, groups of individuals or teams.)
  • The license is non-negotiable.

Information required to complete the license:

  • Legal company name
  • State/country of incorporation
  • Type of corporation
  • Principal address of corporation
  • The "field of use" with which the company plans to use OpenPose
  • Name and title of person who will sign the license
  • Title of the person who will sign the license
  • Address for notices, including name and email of person to be notified

Steps to execute license:

  • ​Press 'Buy" link at the bottom of page
  • You'll be directed to another page to enter contact information
  • You'll then receive a Docusign document with the license
  • Review the information and electronically sign it
  • The signed document will automatically return to CMU, where it will be signed by CMU
  • You'll receive an "Approved" email along with a scan of the fully executed document, and will be given a URL to pay the license fee
  • Pay the license fee
  • Once paid, you'll receive a download link for the code

This technology may not be exported to Cuba, Iran, North Korea, Syria, and Sudan.

Just looking for an academic license?

Operating Systems

  • Ubuntu 14 and 16.
  • Windows 8 and 10.
  • OpenPose has also been used on Windows 7, Mac, CentOS, and Nvidia Jetson (TK1 and TX1) embedded systems. However, we do not officially support them at the moment.

Files

File Name Description
OpenPose license template None Download

Offerings

Name Price
Commercial License, yearly renewal 25,000.00 USD Buy