On November 13th we hosted our first Deep Learning Computer Vision Practitioners’ evening at Popsa, along with co-host Charlotte Bax from Captur.

The idea came about because we realised that there is still very little guidance on advanced deep learning computer vision issues online. There are fantastic online resources to help you get started such as Pyimagesearch.com, and Stackoverflow is always a great help on some of the common problems, but, as practitioners in this rapidly evolving field, we often come across new problems that only a few others have encountered before who may have not yet had the opportunity to share their potential solutions online.

So, we thought it would be a great idea to bring practitioners together to discuss the complex issues we are all encountering, share solutions, and generally encourage and inspire each other.

We had a wide range of attendees, from established businesses to startups to academics, and an inspiring list of speakers including Merve Alanyali from LV, Łukasz Kopeć from Popsa, Charlotte Bax from Captur, Lukas Kruitwagen from the University of Oxford and Marco Paldini from Greyparrot. The topics we talked about varied from challenges in getting labelled data at scale to best practices in training Convolutional Neural Networks.

Future sessions will not only include presentations by practitioners on what they are working on, but also deep dive sessions on specific issues we face with computer vision applications.

If you are a practitioner in this space and want to join us next time, email dlcv@popsa.com to join the mailing list.