Modeling

How important is subject similarity for transfer learning?

Using transfer learning [https://blog.roboflow.com/a-primer-on-transfer-learning/] to initialize your computer vision model from pre-trained weights rather than starting from scratch (initializing randomly) has been shown to increase performance

Zero-Shot Content Moderation with OpenAI's New CLIP Model

When creating a platform on which people can create and share content, there’s often a question of content moderation [https://besedo.com/resources/blog/what-is-content-moderation/]. Content moderation can mean

What is Embedded Machine Learning?

Machine learning – the software discipline of mapping inputs to outputs without explicitly programmed relationships – requires substantial computational resources. Traditionally, this limits where machine learning models can run to very powerful

ELI5 CLIP: A Beginner's Guide to the CLIP Model

You may have heard about OpenAI's CLIP model [https://openai.com/blog/clip/]. If you looked it up, you read that CLIP stands for "Contrastive Language-Image Pre-training.

Liquid Neural Networks in Computer Vision

Excitement is building in the artificial intelligence community around MIT's recent release of liquid neural networks. The breakthroughs that Hasani and team have made are incredible. In this post, we will discuss the new liquid neural networks and what they might mean for the vision field.

How to Train and Deploy a License Plate Detector to the Luxonis OAK

In this post, we will leverage Roboflow and the Luxonis OAK to train and deploy a custom license plate model to your OAK device.

Using Computer Vision to Win at Duck Hunt

Can we use object detection to automate identifying moving objects on a screen? Abhinav Mandava leverages Roboflow to create an aimbot (which automates aiming and firing for the player) for Duck Hunt.

A Primer on Transfer Learning

Transfer learning [https://blog.roboflow.com/what-is-transfer-learning/] is a machine learning (ML) technique where knowledge gained during training a set of problems can be used to solve other related problems.

How to Try CLIP: OpenAI's Zero-Shot Image Classifier

Earlier this week, OpenAI dropped a bomb on the computer vision world.

Football, Kaggle, Roboflow: Using Computer Vision to Tackle Helmet Safety

If you're searching for a dataset to use or are looking to improve your data science modeling skills, Kaggle [https://www.kaggle.com/] is a great resource for

How to Train Scaled-YOLOv4 to Detect Custom Objects

Object detection technology advances with the release of Scaled-YOLOv4. This blog is written to help you apply Scaled-YOLOv4 to your custom object detection task, to detect any object in the world, given the right training data.

Apple's M1 is up to 3.6x as fast at training machine learning models

We compared the Apple M1 chip to the Intel Core i5 chip on an object detection task using Create ML.

5 Strategies for Handling Unbalanced Classes

Suppose you're trying to teach an alien – like one of the crewmates from the wildly popular game Among Us [http://www.innersloth.com/gameAmongUs.php] – to tell the

Scaled-YOLOv4 is Now the Best Model for Object Detection

(based on Microsoft COCO benchmarks) The object detection space remains white hot with the recent publication of Scaled-YOLOv4 [https://arxiv.org/abs/2011.08036], establishing a new state of the

What is Active Learning? The Ultimate Guide.

In this guide, we discuss what active learning is, types of active learning, and walk through an example of active learning in practice.

Google Researchers Say Underspecification is Ruining Your Model Performance. Here's Five Ways to Fix That.

We read that Google underspecification paper so you don't have to.

YOLOv4 - Ten Tactics to Build a Better Model

The YOLO v4 repository is currently one of the best places to train a custom object detector, and the capabilities of the Darknet repository are vast. In this post, we discuss and implement ten advanced tactics in YOLO v4 so you can build the best object detection model from your custom dataset.