Electronics and Communication

PhasorLab

A computer vision solution that augments proprietary high-precision network-based positioning

Overview

PhasorLab uses High-Precision Synchronization Technology, Hyper Sync Net. It can maintain time synchronization of better than one nanosecond and frequency synchronization of better than one ppb, which is key to achieving a centimeter-level target tracking accuracy for network-based positioning systems suitable for indoor & outdoor applications.
Ajackus Partnered with Phasorlab developed a custom model that predicts a 2d image depth up to 10 meters.

Project Name

Phasorlab

Services

Web App development

Domain

domain

Electronics and Communication

Phasorlab

Problem

PhasorLab wanted a computer vision solution that augments proprietary high-precision network-based positioning. They also wanted this solution to be usable with a single CCTV camera.

Process

We divided requirements into two major modules,

  1. Body or Object Detection Module - Find body shapes in a steady stream of images coming at an interval x.Body or Object Detection Module - Find body shapes in a steady stream of images coming at an interval x.
  2. Depth Detection or Distance Measurement Module - Calculate the distance of objects from the camera given known area map.

Description

For body detection, we had to detect a human in an image. We came across many helpful libraries that would help us do this. Mask RCNN is a useful object and instance segmentation library. It gives us good object detection results, and we could pick out the humans detected in the images.

For depth detection, we needed to determine the distance of the human from the camera. Thus depth detection needed to be done on humans from the image input. After a good amount of research on the depth detections algorithms available online, we decided to use the DenseDepth algorithm model for depth detection.

Results

Once the object detection and depth detection is on the image, we extract only the humans' depth from the image. To do this, we used the instance segmentation of the Mask RCNN object detections & pulled the depth detection of the human in the picture.

You may also like

Let's Work Together

Get in touch to let us know what you’re looking for, and one of our solution architects will get back to you soon.

Contact Us