Practical Image Data Augmentation Methods for Training Deep Learning Object Detection Models


Date:  Tuesday, September 22, 2020

Start Time: 9:30 am

End Time: 10:30 am

Data augmentation is a method of expanding deep learning training datasets by making various automated modifications to existing images in the dataset. The resulting increased data diversity can enable a more accurate and robust model without the need to manually obtain more images. In this presentation, we explore practical methods of image data augmentation for training object detection models. We will also show how to create an augmented dataset of 50,000 unique images with labeled bounding boxes in a few hours using a short Python script.

Track

Session Speakers

  • Evan is a full-time photonics hardware engineer at MKS Instruments and a part-time computer vision engineer at his company EJ Technology Consultants. His interest in computer vision started with hobby projects on the Raspberry Pi, and it has evolved into helping companies create products with real-time vision capability. His YouTube channel, Edje Electronics, focuses on democratizing access to computer vision technology by providing easy-to-follow projects and tutorials on using TensorFlow, OpenCV and more. He received a B.S. in Electrical Engineering from Montana Tech, and an M.S. in Electrical Engineering with a focus on ASIC design from Washington State University.


Share

Practical Image Data Augmentation Methods for Training Deep Learning Object Detection Models