Using a pre-trained object detection model to detect people in a selected Region of Interest (RoI).
DreamAI is an under development deep learning library written in Pytorch and inspired by FastAI. This notebook is not a tutorial on DreamAI, it is just an example of how its model can be used for some use case. Stay tuned for a detailed overview/tutorial of DreamAI as well as other examples.
You can find the library here: https://github.com/fzaidi2014/dreamai
Any kind of feedback is much appreciated.
We will use our object detection model that we trained using DreamAI in https://github.com/HamzaFarhan/COCO-Object-Detection-using-DreamAI
The user will select an RoI in a video and then a bounding box will be made around any person who enters that region.
Download or clone DreamAI from the link provided above.
Download or clone this repository, make sure you set up the paths correctly, and the notebook should run smoothly.