• 6 min read
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.
• 6 min read
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.
• 3 min read
Hands on with the Roboflow Infer Web Application Interface Builder
SUMMARY
After training a model with Roboflow Train, the Example Web App gives an immediate browser-based interface to test predictions: upload an image or provide a URL, adjust confidence thresholds and bounding box overlap, filter classes, and view results as an annotated image or raw JSON. Because the app
• 3 min read
Evaluating Object Detection Models with mAP by Class
SUMMARY
Overall mAP scores can mask weak spots in a multi-class object detection model: a 98% aggregate hides the fact that one class is dragging performance down. Roboflow Train addresses this by surfacing per-class mAP on both the validation and training sets after every training run, so you
• 5 min read
How This Fulbright Scholar is Using Computer Vision to Protect Endangered Species
SUMMARY
Kasim Rafiq, a Fulbright Scholar and National Geographic Explorer at UC Santa Cruz, is building a vehicle-mounted computer vision system that automatically detects and photographs wildlife during safari game drives, then uses those detections for species population surveys at a fraction of the cost of traditional field methods.
• 2 min read
Train Test Split Guide and Overview
SUMMARY
Splitting a dataset into training, validation, and test sets is a core practice for detecting overfitting and ensuring a computer vision model generalizes beyond its training examples. Roboflow defaults to a 70/20/10 split at upload time but gives you direct control: you can assign individual image batches
• 31 min read
Software Engineering Daily Podcast
Roboflow co-founder Brad Dwyer was a guest on the Software Engineering Daily podcast. Listen on your favorite podcast app (Apple Podcasts, Spotify, Overcast, Stitcher), or see the full transcript below.
Software Engineering Daily: Roboflow: Computer Vision Models with Brad Dwyer on Apple PodcastsTraining a computer vision model is not
• 2 min read
Introducing an Improved Shear Augmentation
Today, we introduce a new and improved shear augmentation. We'll walk through some details on the change, as well as some intuition and results backing up our reasoning.
• 2 min read
Fighting Wildfires with Computer Vision
SUMMARY
Abhishek Ghosh, a graduate student at Texas A&M working with AI For Mankind, trained an EfficientDet D0 computer vision model to detect early smoke from forest fires using images from weather towers. Using Roboflow Pro, he applied advanced augmentations and a static crop along the horizon line
• 6 min read
How Tesla Teaches Cars to Stop
SUMMARY
Tesla Senior Director of AI Andrej Karpathy's CVPR 2020 talk reveals that even a seemingly straightforward computer vision task like stop sign detection involves dozens of edge cases: occluded signs, non-standard mounting, signs that do not require stopping, and conditionally active signs. Tesla handles this by
• 1 min read
Roboflow Wins $15,000 Award
Roboflow is honored to be named as one of Iowa's most promising startups and one of the top 3 winners of the Pappajohn Entrepreneurial Venture Competition. The annual awards honor companies with innovative technology that have grown quickly.
Since launching in January, Roboflow has helped over 8,000
• 3 min read
Introducing Grayscale and Hue/Saturation Augmentations
SUMMARY
Roboflow has added three color-based augmentation options, grayscale, hue, and saturation, to help models generalize to objects that appear under different lighting conditions or in varied color environments. Grayscale augmentation randomly converts a configurable percentage of training images to single-channel output, reducing the model's reliance
• 3 min read
Training Robots to Identify Other Robots
SUMMARY
Amitabha Banerjee trained a computer vision model to teach his Anki Vector robot to detect other robots, addressing a real gap since the Vector's onboard AI handles speech and human recognition but not robot-to-robot identification. He collected 303 images from the robot's own
• 1 min read
Elisha Odemakinde Hosts Roboflow for a Fireside Chat
Recently, Roboflow machine learning engineer Jacob Solawetz sat down with Elisha Odemakinde, an ML researcher and Community Manager at Data Science Nigeria, for a Fireside chat.
During the conversation, Jacob walks through his journey from his career in financial trading to artificial intelligence, a move from natural language processing to
• 4 min read
Deploy YOLOv5 to Jetson Xavier NX at 30FPS
Learn how to deploy a YOLOv5 model to a Jetson Xavier NX edge device.
• 7 min read
Use Resnet34 for Image Classification
SUMMARY
This tutorial walks through training a custom image classification model on the ResNet34 backbone using the fastai library and PyTorch, starting from an ImageNet pretrained checkpoint and fine-tuning it on a new dataset via transfer learning. The post uses a public flower classification dataset as a worked example,
• 2 min read
Explore Images, Annotations, and Metadata Easier
SUMMARY
Roboflow's updated image preview interface lets users view annotations full-screen, toggle individual annotation groups on and off by hovering, and surface image metadata including the original filename, camera used, modification date, and which split the image belongs to. For dataset versions with augmentations applied, the preview
• 4 min read
Fast.ai v2 Released - What's New?
SUMMARY
Fastai v2 brings a restructured three-layer API built on PyTorch, making it practical to move between computer vision and NLP tasks with minimal code changes. Key additions include a flexible infinite training loop, GPU-accelerated per-batch augmentation, and companion libraries (fastcore, fastscript, fastgpu) that extend Python and
• 2 min read
Roboflow Joins Y Combinator's Summer 20 Batch
Roboflow is enabling any developer to use computer vision (without being a machine learning expert).
Computer vision is the first technology that fundamentally allows us to rewrite human-computer interaction. Until now, when using computers to interact with the world, humans would assess their environment, provide commands to a computer,
• 1 min read
Video: How to Use Amazon Rekognition Custom Labels with Roboflow
SUMMARY
This webinar recording from Roboflow and AWS Premier Consulting Partner Onica covers the full pipeline for training and deploying a custom mask detection model using Roboflow for dataset preparation and AWS Rekognition Custom Labels for training and hosting. The session goes end to end from raw images to a
• 8 min read
How to Train a Custom Mobile Object Detection Model (with YOLOv4 Tiny and TensorFlow Lite)
Learn how to train a custom mobile object detection model with YOLOv4 tiny and TensorFlow Lite.
• 3 min read
The crazy story of how we got our .com domain
For almost a year, Roboflow (our computer vision dataset management tool) has lived at roboflow.ai. It's served us well but we have always lusted after the dot com. No longer! We managed to acquire roboflow.com from the giant German company that owned it since 2001.
The
• 4 min read
Using Computer Vision to Find Brands in YouTube Videos
SUMMARY
Software engineer Charles Herring built a browser-based computer vision model that detects power tool brands appearing in DIY YouTube videos, giving companies a way to measure unpaid exposure and identify sponsorship opportunities. Roboflow replaced Google Cloud AutoML as his data management layer, providing dataset-level class distribution statistics
• 14 min read
Deploy A Custom Model to the Luxonis OAK - A Quickstart Guide
This guide will take you the long distance from unlabeled images to a working computer vision model deployed and inferencing live at 15FPS on the affordable and scalable Luxonis OpenCV AI Kit (OAK) device.
Stay Connected