Evaluating the Performance of Training in YOLO Deep Learning Networks with Insignificantly Small Dataset

Abstract

In the last few years, Deep Learning is the one of the top research areas in academia as well as in industry. Every industry is now looking for a deep learning-based solution to the problems in hand. As a researcher, learning “Deep Learning” through practical experiments will be a very challenging task. Particularly, training a deep learning network with huge amount of training data will make it impractical to do this on a normal desktop computer or laptop. Even a small-scale application in computer vision using deep learning techniques will require several days of training the deep network model on a very higher end GPU clusters or TPU clusters – that makes impractical to do that research on a conventional laptop.

In this work, we address the possibilities of training two versions of YOLO deep learning networks with an in significantly small dataset. Since we are going to design a prototype drone detection system with two different network models which are dealing with single class classification problem, we hereby try to train the deep learning networks only with few drone images (2 images only) and compare their performance in terms of mean average precision (mAP) and other suitable metrics.

The arrived results prove the possibility of training deep learning network with very few images (or any data). According to the results, YOLOv3 performed better than VOLOv2 and proves the possibility of training a deep learning network with insignificantly low number of images