This repository contains software related to human sensing.
It is currently under major overhauling and all previous software + documentation has been kept in the under-review branch and will soon be transferred to master. In the meanwhile, if parts of the code is required please use the following git command if you already have a version pf the repo: git checkout under-review otherwise directly clone the under-review branch: git clone -b under-review https://github.com/robotology/human-sensing.git
Project updates can be found here
The human sensing repository contains the following directories and modules:
faceLandmarks: This yarp module module is configured to find human faces that are looking more or less towards the camera. It makes use of the dlib library in order to extract the facial landmarks and uses the Histogram of Oriented Gradients (HOG) feature combined with a linear classifier, an image pyramid, and sliding window detection scheme.
yarpOpenPose: This yarp module is configured to estimate human pose based on OpenPose, a library for real-time multi-person keypoint detection and multi-threading. It makes use of Caffe and OpenCV.
-
googleVisionAI: This yarp module is configured to perform image classification using the Vision AI google service, which makes use of AutoML Vision or use pre-trained Vision API models to detect emotion, understand text, and more.
In order to use the module, click here and follow the instructions explained in the Step 3.
At the present time each directory is independent from the others and must be compiled and set up separately. While we provide specific usage instructions for each one in separate files, we will soon integrate all modules as a unique build procedure.
In order to compile and run modules in this repository you will need to install the following dependencies. Installation are standards and instructions can be found in the following links.
Please find all the related and detailed installation instructions at the following link.
Please find all the related and detailed installation instructions at the following link.
Online documentation is available here: http://robotology.github.com/human-sensing.
Material included here is Copyright of iCub Facility - Istituto Italiano di Tecnologia and is released under the terms of the GPL v2.0 or later. See the file LICENSE for details.