There exist a plethora of articles on quantization, but they generally cover only the surface level theory or provide a simple overview. In this article, I’ll explain how quantization is actually implemented by deep learning frameworks.
Before getting into quantization, it’s good to understand the basic difference between two key concepts: float (floating-point) and int (fixed-point).
I’m by no means a developer, engineer, or coder of any kind. Though I do read a lot of tutorials that include code implementations, and occasionally peruse repos to explore what folks are building, the most I’ve ever done is add some CSS style elements to a button within an HTML file.
So on the surface, the idea of me being able to build a cross-platform, mobile-ready machine learning model is kind of absurd.
Developers using Fritz AI can now track the body positions of people in images and video with Pose Estimation.
Deploying machine learning-based Android apps is gaining prominence and momentum with frameworks like TensorFlow Lite, and there are quite a few articles that describe how to develop mobile apps for computer vision tasks like text classification and image classification.
But there’s very much less that exists about working with audio-based ML tasks in mobile apps, and this blog is meant to address that gap — specifically, I’ll describe the steps and code required to perform audio classification in Android apps.
Continue reading “Audio Classification in an Android App with TensorFlow Lite”
For farm owners in developing countries, having access to the right education and tools to aid in crop production is important to improving their livelihoods. PlantVillage, a research group out of Pennsylvania State University, leverages machine learning in areas of limited connectivity to diagnose plant disease and advise farmers on how to treat their crops.
Google’s Colab is a truly innovative product for machine learning. It enables machine engineers to run Notebooks and easily share them with colleagues. Another key advantage is access to GPUs and TPUs.
In this piece, we’ll highlight some of the tips and tricks mentioned during this year’s TF summit. Specifically, these tips will help you in getting the best out of Google’s Colab.
Continue reading “Top 10 Tricks for TensorFlow and Google Colab Users”
In the last piece in this series on developing with Flutter, we looked at how we can implement text recognition using ML Kit, which belongs to the Firebase family.
In this sixth installment of the series, we’ll keep working with ML Kit, this time focusing on implementing image labeling. The application we build will be able to label the contents of an image and provide a confidence level.
Continue reading “Image Labeling in Flutter Using Firebase’s ML Kit”
Among the new capabilities in Lens Studio 3.4, one that might have flown a bit under the radar—especially with the really impressive 3D Body Tracking, Full-Body Segmentation, and enhanced Finger Tracking taking a lot of the spotlight—is the new Asset Library.
Put simply, it’s a new collection of ready-to-use assets, provided by the Snap team and by a few select Official Lens Creators. If you want to get started with a range of different capabilities inside Lens Studio, but don’t want to go elsewhere to build 3D objects, ML models, materials, and more, this new library is where you should start.
Continue reading “Experimenting with Lens Studio’s Asset Library”
This blog is a continuation of our series comparing on-device OCR libraries. Here’s a similar comparison on iOS devices.
Summarizing the results of our comparison on iOS devices, Firebase’s ML Kit was leading by a solid margin against Tesseract OCR. However, outcomes weren’t very similar on Android phones. For this particular test, we used Samsung Galaxy J7 with 2GB RAM and 32GB Memory. Different hardware may result in different results.
Continue reading “Choose the Right On-Device Text Recognition (OCR) SDK on Android Using DeltaML”
A month back I wrote a post that introduced an open-source package react-native-text-detector. In that tutorial, we built a simple Business Card app using the same package. Package used different libraries for detection on Android and iOS. It used Firebase’s ML Kit on Android and Tesseract OCR along with Core ML on iOS. The major reason for this is mentioned here.
Since the conflict between ML Kit and React Native on iOS was resolved, we now have to select one proper solution for achieving this task on both platforms. In order to make an informed decision, we had to analyze the performance of both of these libraries on both platforms.
Continue reading “Comparing iOS Text Recognition SDKs Using Delta”
