The computer vision space is experiencing a dramatic technology shift, with a growing number of engineers using deep learning and convolutional neural networks to develop the next generation of computer vison applications. And when it comes to designing, training, evaluating and deploying neural networks, TensorFlow is stealing the show.

Since its introduction two years ago, Google’s open source TensorFlow framework has emerged as the most popular deep learning design framework for computer vision applications. That’s why we decided to focus on TensorFlow for our 2018 series of training classes: Deep Learning for Computer Vision with TensorFlow.

The single-day course provides a hands-on introduction to the practical use of Google’s TensorFlow for computer vision applications—in particular, object recognition. We start with a few simple examples (e.g., linear regression and logistic classification) to show you the basics, then proceed to multi-layer deep networks and convolutional neural networks for object recognition. After that, we show you how to do “transfer learning”, which allows you to quickly retrain an existing network for your particular tasks. Finally, we show you the latest developments in TensorFlow for computer vision, including using TensorFlow on embedded platforms.

The three-day program provides a deep dive. The first day covers the basics of machine learning such as linear regression, logistic classification, multi-class classification, and how to construct deep networking—all using standard Python techniques, so you can really understand what’s happening behind the scenes and how best to optimize a network. The second day focuses on how to use the TensorFlow framework for linear regression, logistic classification, and deep networks. The last day includes visualizing learning, convolutional neural networks, and transfer learning. We also cover object detection and how to run networks on embedded and mobile platforms.

And all classes offer first-rate cloud infrastructure sponsored by IBM and Nimbix, leveraging IBM PowerAI on the Nimbix cloud platform.

We’re hosting classes in San Jose, Seattle, New York City, Boston, and Santa Clara – don’t miss out! You can find them all on the Events page.


Jeff Bier, Founder
Embedded Vision Alliance

Join The Discussion

Your email address will not be published. Required fields are marked *