Jaiaaryan / computer_vision

C/C++/Python based computer vision models using OpenPose, OpenCV, DLIB, Keras and Tensorflow libraries. Object Detection, Tracking, Face Recognition, Gesture, Emotion and Posture Recognition

Gesture, Emotions, Posture and Face Recognition using OpenPose/DLIB

Purpose of this work is to demonstrate few state of art computer vision applications using OpenPose/DLIB libraries.

Acknowledgements:

This github repository work is greatly inspired and have used code, concepts presented in the following github repositories:

• DLIB: Modern C++ toolkit for computer vision and other machine learning.
• Kerasify : Small library for running Keras models from a C++ application.
• OPENCV: Open Source Computer Vision Library.
• OPENPOSE: A Real-Time Multi-Person Keypoint Detection And Multi-Threading C++ Library.

Thanks to Dr.Michael Rinehart, Chief Scientist at Elastica for his mentorship and guidance through the project.

Operating systems (supported):

• Ubuntu 16.04
• Nvidia Jetson TX2

Requirements:

• NVIDIA graphics card with at least 1.6 GB available (the nvidia-smi command checks the available GPU memory in Ubuntu).
• At least 2 GB of free RAM memory.
• Highly recommended: cuDNN and a CPU with at least 8 cores.

Install, Compile and Run:

• install_compile_and_run (INSTALLATION IS MUST..)

Design:

• Software design

Demo:

Gesture recognition:

Emotions recognition:

Pose recognition:

DLIB Face recognition: