Object detection(YOLO) on Nvidia jetson TX2

Tensorflow and keras implementation of YOLO algorithm using the on-board camera of TX2.

Demo Link:https://youtu.be/rHyBtfIDGOc

Getting Started

The neural network was trained on Nvidia Titan X GPU.This model was later used with nvidia Jetson TX2 Board. Opencv was used to capture images .

Prerequisites

1.Python 3.5
2.Tensorflow 1.5
3.Keras
4.Scikit Learn
5.Open CV 3.4.1

Theory

I.A YOLO algorithm

							  Broad Overview
								|
						    Reduce Dimesions of image
								|
		                 probablity of box | box co-rdinates | Probability of each class
								|
							  hard threshold
								|	
							Non max-suppression
								|
							      output

I.B Non-max Supression

II. Open CV

1.Open CV installation for ubuntu is pretty standard and done in same way as shown on their website.
2.Installation of Open CV on Jetson TX2 is a bit diffrent.Jetson Hacks has a pretty cool script on his Git repo.This enable open CV on GPU on the Jetson.

III.Running on Jetson TX2

This was done using pre-trained model by darknet.Follow the instruction as follows.

Then run the test_on_cam.py file.
Uncomment if needed to run on Video.Add video to data folder in .avi format.

Download the dataset: https://www.dropbox.com/s/nli1ne8hzkzsyt6/NFPAdataset.zip?dl=0

Results

Yolo-tiny is a lightweight option with pretty descent accuracy and does not require huge computation resources.Athough a correct and accurate implementation is th YOLO designed above.

Authors

• Siddharth Bhonge - Parser /Model - https://github.com/siddharthbhonge

Acknowledgments

• Andrew Ng | Deeplearning.ai
  

• Joseph Redmon, Santosh Divvala, Ross Girshick, Ali Farhadi - You Only Look Once: Unified, Real-Time Object Detection (2015)
  

• Joseph Redmon, Ali Farhadi - YOLO9000: Better, Faster, Stronger (2016)
  

• Allan Zelener - YAD2K: Yet Another Darknet 2 Keras
  

• The official YOLO website (https://pjreddie.com/darknet/yolo/)