Deevia Human Pose Estimator

Deevia Human Pose Estimator

Deevia HumanPose Estimator analyzes human activities in Industrial Assembly for improving safety (Bent, Arms stretched, Squat) and productivity (Working, Not Working, Sitting, Walking, talking over Phone/Reading/Observing) in real-time.The solution is a real-time inference system running on CPU using the OpenVINO framework for Human Pose Estimation acting as features. Intel’s OpenVINO Human Pose Estimator is also used to get Joint Positions (skeletons) to train Machine/Deep Learning-based models to identify activities (Walk, Work, Arms Stretched, Squat, Bend, etc.). Single frame analysis/models are sufficient for activities like Arms Stretched, Squat, Bend. Multi-frame analysis/models are required for activities like Walking, Working, Talking to Others, etc.

*Please note that member solutions are often customizable to meet the needs of individual enterprise end users.

CONTACT COMPANY

SOLUTION FEATURES

  • Real-time camera-based solution
    The solution is a real-time inference system running on CPU using the OpenVINO framework for Human Pose Estimation acting as features.
  • OpenVINO Human Pose Estimator
    Intel’s OpenVINO Human Pose Estimator is used to get Joint Positions (skeletons)
  • Multi-person pose estimation
    Real-time multi-person system to jointly detect human body key-points (18 key-points) on single images

CATEGORIES

Worldwide Intel® Xeon Scalable Hybrid Cloud On-premise (Private Cloud, Other) Caffe Cross-Industry Manufacturing Models cant be re-trained - Inference only AI Software/SaaS Linux Windows Intel® Distribution for Python Intel® Distribution of OpenVINO™ toolkit Proprietary Image/Object Detection/Recognition/Classification Video Surveillance and Analytics Deep Learning