Posts Written by Jacob Solawetz

Jacob Solawetz

Founding Engineer @ Roboflow - ascending the 1/loss

What is YOLOv6? The Ultimate Guide.

In this guide, we discuss what YOLOv6 is, how the model works, and the architectural improvements made in YOLOv6 over its predecessors.

What is YOLOv4? A Detailed Breakdown.

In this guide, we discuss what YOLOv4 is, the architecture of YOLOv4, and how the model performs.

What is YOLOv7? A Complete Guide.

In this guide, we discuss what YOLOv7 is, how the model works, and the novel model architecture changes in YOLOv7.

Scaling up ViT Image Classification on Gaudi2 HPUs

Training large image transformers comes with heavy compute requirements - enter the Gaudi2 from Habana Labs, an Intel company.

CVPR 2023 Highlights

Three members of the Roboflow team attended CVPR this year. Read our highlights from the conference and what trends we noticed.

What is Segment Anything Model (SAM)? A Breakdown.

Learn how Meta Research's new Segment Anything Model works to achieve high performance on image segmentation tasks.

Preview: Roboflow + GPT-4

Roboflow and GPT-4 will be even more powerful when used in conjunction. In this post we preview some of the new features that will be coming to Roboflow in the coming weeks.

Speculating on How GPT-4 Changes Computer Vision

OpenAI released GPT-4 showcasing strong multi-modal general AI capabilities in addition to impressive logical reasoning capability. Are general models going to obviate the need to label images and train models?

What is YOLOv8? The Ultimate Guide. [2024]

The field of computer vision advances with the newest release of YOLOv8, setting a new state of the art for object detection and instance segmentation.

HPU vs GPU - Benchmarking the Frontier of AI Hardware

When you are training machine learning models, it is essential to pick hardware that optimizes your models performance relative to cost. In training, the name of the game is speed per epoch – how fast can your hardware run the calculations it needs to train your model on your data.

📸 Roboflow 100: A Multi-Domain Object Detection Benchmark

Roboflow 100 (RF100) is a crowdsourced object detection benchmark. The dataset consists of 100 datasets, 7 imagery domains, 224,714 images, and 829 class labels with over 11,170 labeling hours.

CVPR 2022 - Best Papers and Highlights

In this post, we take the opportunity to reflect on the computer vision research landscape at CVPR 2022 and highlight our favorite research papers and themes.

Train and Deploy YOLOS Transformer On a Custom Dataset

In this post, we showcase training and deploying YOLOS end to end, from labeling your data, to training your model, to deploying your model on AWS for inference.

What is Instance Segmentation? A Guide with Roboflow.

We are excited to release support for instance segmentation projects on Roboflow. Instance segmentation allows your computer vision model to know the specific outline of an object in an image, unlocking new use cases for Roboflow in your application.

How to Train Detectron2 for Custom Instance Segmentation

A walk through on how to train Detectron2 to segment your custom objects from any image by providing our model with example training data.

What is CVAT (Computer Vision Annotation Tool)?

Learn how to annotate images in CVAT, an open-source, web-based tool for labeling data for object detection, segmentation, classification, and other tasks.

Introducing New and Improved Roboflow Train

Over the last year, thousands of custom computer vision models have been trained with Roboflow Train [] and millions of inferences have been made via Roboflow

What to Think About When Choosing Model Sizes

When training any machine learning model, you must trade off inference speed for accuracy. Larger models with more parameters are uniformly more accurate, and smaller models with fewer parameters are uniformly faster to infer.

Introducing the Roboflow Universe Dataset Research Internship

We are seeking an intern who would be interested in first-authoring the paper that introduces the Roboflow Universe Datasets.

SageMaker Studio Lab vs Google Colab

Recently, AWS released SageMaker Studio Lab, its competitor service to Google Colab. I dove into comparing Google Colab to Studio Lab and here is what I found.

Introducing the Roboflow Inference Widget

Now, you can easily test any model that has been trained with Roboflow Train by dragging an image file onto your dataset version page.

Florence: A New Foundation for Computer Vision

Microsoft Research recently released the foundational Florence model, setting the state of the art across a wide array of computer vision tasks.

What is YOLOS? What's New in the Model?

YOLOS - You Only Look At One Sequence is the newest, and potentially most impactful, iteration on the YOLO family of object detection models.

YOLOv5 v6.0 is here - new Nano model at 1666 FPS

With the v6.0 release, YOLOv5 further solidifies its position as the leading object detection model and repository.

How to Safely Install OpenCV on the Mac M1

Installing OpenCV on the M1 safely is difficult because the M1 operates on an arm64 architecture and most of your python libraries are compiled for amd64. Open this guide to avoid your otherwise inevitable demise.