Blog

Editor’s Note: For an in-depth look at SnapML, Lens Studio’s new framework for leveraging custom machine learning models inside Snapchat Lenses, download our free comprehensive guide.

OLC Kaden (guccikaden) is well on his way to 100k subscribers on Snapchat. The young creator tends toward vintage-inspired lenses that make this Millennial yearn for the days of Lisa Frank stickers and Y2K. With so many fun, aesthetic Lenses, it’s no surprise Kaden is making waves in the Lens creation world. I had a quick chat with the Lens maker to learn more.

Continue reading Snapchat Lens Creator Spotlight: Kaden

In a previous tutorial, I walked through how to train a style transfer model with Fritz AI and implement it inside Lens Studio to create a custom artistic Snapchat Lens:

While this was a really cool first step, it left me itching to push things just a bit further. Specifically, I wondered what it would take to manipulate where in the camera scene the style transfer model is actually applied. For instance, could I apply the artistic style only to the background and not the human subject, creating a semi-virtual background?

Continue reading “Combining Style Transfer with Background Replacement in Lens Studio with Fritz AI and SnapML”

Facebook’s AR Studio is used for building augmented reality experiences targeted at the Facebook and Instagram platforms.

In this tutorial, let’s look at how you can use this software to build an effect that responds to someone’s face. To achieve this you will take advantage of AR Studio’s Face Tracker, Face Mesh, and Texture.

Continue reading How to Create a Face Tracking Effect in Facebook’s AR Studio

AR camera filters that can be accessed via social media apps like Facebook and Instagram are referred to as “social” applications. Inside the camera feature of these applications, there are a variety of amusing and attractive filters. Facebook and Instagram both have AR capabilities.

Spark AR Studio can be used to modify the appearance of people’s faces. You may also give the face and the whole scene a finishing touch. You will learn how to do some minor face smoothing and lighting effects in this blog.

Continue reading “Face Changing Effect In Spark AR”

NLP is a hot topic right now in the deep learning and machine learning communities, thanks to the incredible development of Transformer-based neural networks like BERT and GPT-2.

Natural language processing (NLP) gives programs and applications the ability to understand spoken and written human language.

Developing NLP applications is difficult because traditionally, computers are designed for humans to interact with them in precise, unambiguous, and highly-structured programming languages, or by using a limited number of set spoken commands.

Continue reading “Sentiment Analysis iOS Application Using Hugging Face’s Transformers Library”

It’s seems that, in the last decade, people increasingly have turned to social networks to express their opinions on everything from everyday life to politics and even businesses. This has been an important source — for businesses — to probe the market and get a clear picture of the needs and directions that influence it.

For me, the social media that seems the most active in the area of expressing opinions is probably Twitter.

Continue reading “Processing Tweets Using Natural Language and Create ML on iOS”

Machine learning has gained plenty of momentum recently, and with Google’s announcement of TensorFlow Lite, it’s never been easier to start with incorporating machine learning directly in your mobile apps. Mobile machine learning has become a highly sought after skill, and after being introduced to TensorFlow in Google’s annual keynote, I realized that it could solve real world problems, like analyzing complex CT scans to detect tumors or reading handwritten letters to the visually impaired. With that, I decided to jump in.

Continue reading “Tutorial: Troubleshooting TensorFlow Lite on Windows 10”

Following up on my earlier blogs on running edge models in Python, this fifth blog in the series of Training and running Tensorflow models will explore how to run a TensorFlow Lite image classification model in Python.

Continue reading “Running TensorFlow Lite Image Classification Models in Python”

In gradient boosting, predictions are made from an ensemble of weak learners. Unlike a random forest that creates a decision tree for each sample, in gradient boosting, trees are created one after the other. Previous trees in the model are not altered.

Results from the previous tree are used to improve the next one. In this piece, we’ll take a closer look at a gradient boosting library called CatBoost.

Continue reading “Fast Gradient Boosting with CatBoost”