Blog

Matt Brems

Growth Manager @ Roboflow. Previously solved data science problems across finance, education, politics, and more. Passionate about teaching and empowering others to accomplish more.

Latest Posts by Matt Brems

Handling unbalanced classes

3 Jan 2025 • 8 min read

5 Strategies for Handling Unbalanced Classes in Machine Learning

SUMMARY Class imbalance causes models to underperform on minority classes because the training distribution does not reflect the real-world distribution of what matters. Five strategies address this: gathering more data from underrepresented classes, generating synthetic samples through augmentation, undersampling the majority class via random selection, oversampling the minority class

CLIP: Connecting text and images

1 Sep 2024 • 15 min read

What is CLIP? Contrastive Language-Image Pre-Training Explained.

CLIP is an open source, multimodal computer vision model developed by OpenAI. Learn what CLIP is in this guide.

The power of image augmentation: an experiment

3 May 2021 • 8 min read

The power of image augmentation: an experiment

SUMMARY Image augmentation increases effective training set size by applying random transformations to existing images, but knowing how much to augment and which transforms to apply depends on your dataset. This experiment tests three variables across three public datasets of different sizes (26, 196, and 665 images of packages, raccoons,

remote work

25 Apr 2021 • 6 min read

How We at Roboflow Do Remote Work

We'd like to think we've worked remotely pretty successfully: we raised a seed round, we've come in first place on the Pioneer leaderboard for a record-breaking 25-weeks in a row, we publish at least five pieces of content each week (check out

using computer vision to keep stock of inventory

18 Apr 2021 • 3 min read

How to Use Computer Vision to Keep Stock of Inventory

Learn how to use computer vision to track inventory in logistics facilities.

docu

6 Apr 2021 • 3 min read

Using Computer Vision to Extract Document Structure

SUMMARY A student in Germany trained a custom object detection model on roughly 1,000 scanned school book pages to identify individual tasks and page numbers, reducing the data that needs to be transmitted during remote learning by sending extracted text instead of full images. The project used Roboflow Annotate

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

28 Mar 2021 • 6 min read

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

Learn how to use the CLIP zero-shot model to moderate visual content.

paperspace gradient

24 Mar 2021 • 1 min read

Webinar: How to Build a Detectron2 Model with Roboflow and Paperspace Gradient

SUMMARY This post announces and links the recording for a joint Roboflow and Paperspace webinar that walks through training a Detectron2 object detection model on a blood cell count dataset. The session covers accessing public image datasets through Roboflow, applying preprocessing and augmentation, and running GPU-backed training notebooks on

cyclists

21 Mar 2021 • 4 min read

Danger Monitoring for Cyclists using Raspberry Pi and Object Detection

SUMMARY This project builds a low-cost cyclist danger monitor using a Raspberry Pi 4, a Google Coral TPU stick for faster inference, and an LED traffic light that warns the rider when cars or bicycles are detected in proximity. The model was trained on three datasets totaling over 6,

ppe

15 Mar 2021 • 3 min read

Using Computer Vision to Detect Personal Protective Equipment

SUMMARY Construction site PPE compliance can be automated with an object detection model that flags safety violations, such as missing hard hats, directly from site images. Data scientist Jaco Lau built this using Roboflow for image annotation, class balancing, and model training, then deployed the detector inside a Streamlit application

opencv

1 Mar 2021 • 3 min read

5 Emerging Trends in Computer Vision Applications from OpenCV's AI Competition

SUMMARY Reviewing over 200 project proposals submitted to the 2021 OpenCV AI Competition, five application patterns stand out across the more than 1,200 competing teams: assisting the visually impaired with spatial awareness, improving workplace safety by identifying hazardous zones, monitoring COVID protocol compliance, and tracking animal health and conservation

How to Use Roboflow and Streamlit to Visualize Object Detection Output

23 Feb 2021 • 8 min read

How to Use Roboflow and Streamlit to Visualize Object Detection Output

SUMMARY This tutorial builds a blood cell detection app that trains an object detection model in Roboflow to identify platelets, red blood cells, and white blood cells, then wraps the model's API in a Streamlit interface so users can upload images, adjust confidence thresholds with a slider, and

earth

22 Feb 2021 • 3 min read

How to Use Google Earth Engine and Python API to Export Images to Roboflow

SUMMARY Google Earth Engine's Python API can pull multispectral imagery from the Sentinel-2 satellite and export it as image files ready to upload directly into Roboflow for training computer vision models on geographic data. The tutorial covers authenticating to both Google Earth Engine and Google Drive, selecting

music

15 Feb 2021 • 3 min read

Using Computer Vision to Improve Disability Access in Music Education

SUMMARY A high school student built the Mehek Box, a tactile rhythm education tool for students with disabilities, and extended it with a computer vision feature that recognizes physical rhythm blocks via an in-app camera. She trained an EfficientDet-D0 object detection model on a custom dataset built from

Using Computer Vision to Make Card Grading Faster and Cheaper

8 Feb 2021 • 4 min read

Using Computer Vision to Make Card Grading Faster and Cheaper

SUMMARY Professional card grading is slow and expensive, and a large portion of the assessment (edge wear, corner wear, centering, surface damage) maps well to computer vision. This guest post describes a prototype that uses OpenCV for edge detection and card alignment checks, then trains a YOLOv5 model on a

duck hunt

31 Jan 2021 • 3 min read

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.

Using Computer Vision to Detect Package Deliveries

19 Jan 2021 • 4 min read

Using Computer Vision to Detect Package Deliveries

SUMMARY A Jetson Nano paired with a $100 IP camera and a model trained iteratively through Roboflow makes a practical, locally-run package detection system that sends a phone notification with a cropped view of the delivered package. The model was built starting from just 15 labeled images, then grown

computer vision to boost city efficiency

4 Jan 2021 • 6 min read

Using Computer Vision to Boost Cities' Efficiency by Reallocating Police Resources

SUMMARY Data scientist Joseph Rosenblum built a computer vision system that turns existing CCTV footage into automated traffic enforcement, detecting violations like crosswalk blockages that officers would otherwise need to monitor in person. The model was trained on the WebCamT dataset of low-resolution New York City webcam footage with

football

28 Dec 2020 • 6 min read

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

SUMMARY This post walks through a Kaggle competition case study focused on detecting American football helmet impacts to help researchers understand conditions that contribute to traumatic brain injury. The authors use Scaled-YOLOv4, selected for its inference speed, and use Roboflow to preprocess and augment the provided training images. A

Tackling the Internet of Things with Roboflow: Object Detection Apps on Android

21 Dec 2020 • 4 min read

Tackling the Internet of Things with Roboflow: Object Detection Apps on Android

Joo chan Kim, PhD student, is developing an object detection application for Android devices that can identify specific IoT sensors by using a custom detection model.

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

15 Dec 2020 • 5 min read

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.

Scaled-YOLOv4 is Now the Best Model for Object Detection

3 Dec 2020 • 8 min read

Scaled-YOLOv4 is Now the Best Model for Object Detection

SUMMARY Scaled-YOLOv4 set a new state-of-the-art result for object detection on the Microsoft COCO benchmark at the time of its publication, surpassing Google's EfficientDet across both the speed and accuracy ends of the performance frontier. The model achieves this by scaling the depth, stages,

How to Run Jupyter Notebooks on an Apple M1 Mac

30 Nov 2020 • 6 min read

How to Run Jupyter Notebooks on an Apple M1 Mac

Learn how to run Jupyter Notebooks on Apple M1 Macbooks.

generating renaissance art computer vision

30 Nov 2020 • 3 min read

Generating Renaissance Art with Computer Vision

SUMMARY Two high school students built a DCGAN during a 24-hour hackathon to generate abstract images in the style of Renaissance paintings, training on roughly 300 images sourced from the web. They used Roboflow to preprocess and scale images to 32x32 pixels, and applied data augmentation (reflections and rotations)

Stay Connected

Get the Latest in Computer Vision First