Human detection software for a defense firm

Tom Potanski

Updated on May 25, 2023 | 2 min read

We made an advanced Computer Vision software for a military company to detect human posture on real-time footage

Detect human contour and center of mass in real-time

Project Overview

One of our customers needed a software solution that detects humans on a video stream in real-life which will be as lightweight as possible to be run on a low cost, small device equipped with an Nvidia GPU.

Performant Nvidia Jetson software

The hardware capabilities of Nvidia Jetson made Deep Learning infeasible for this task. However, we managed to provide a performance-optimized, OpenCV-based algorithm that worked in real-time, with very low latency and with the required precision. To speed up computations, we’ve used the C++ OpenCV library in our Python code – the official Python version didn’t provide access to CUDA algorithms, so we created a dedicated library to provide appropriate bindings.

Further project development

Currently, we are working on extensions to the existing algorithm to enable more comprehensive predictions like real-time body part segmentation and gesture recognition.

Using C++ in many ways:

• Drawing contour over human body
• Leveraging bloom filters and advanced mathematical models
• Real-time video processing
• Handling difficult lightning conditions