Real-Time iOS Camera Video Matting via CoreML
May 5, 2022 1:27 pm by Dmitry Voitekh
In this article, I will share the implementation details of an iOS SwiftUI app with CoreML features. The app captures a front camera stream, applies a state-of-the-art Video Matting model, and outputs a corresponding video sequence without a background (a.k.a. green screen) in real-time.
Part 2: Computer Vision @ GIPHY: How we Created an AutoTagging Model Using Deep Learning
January 11, 2021 5:43 pm by Dmitry Voitekh
This is part two of the GIPHY Autotagging blog post series, where we’ll cover modeling, configuration of our training environment, and share our results. In part one, we outlined our motivation for this product and provided an overview of existing related approaches. Additionally, we described the training and evaluation data we have at hand and […]
Part 1: Computer Vision @ GIPHY: How we Created an AutoTagging Model Using Deep Learning
December 3, 2020 7:42 pm by Dmitry Voitekh
Motivation GIPHY search is powered by ElasticSearch, where GIFs are represented as documents with various features, such as tags and captions. Tags in particular have a great impact on the search performance. Therefore, at GIPHY Engineering, we continuously track and improve their quality with both manual and automated verification. However, for the last couple of […]