Roboflow Blog

8 Aug 2025 • 5 min read

How to use YOLOE for Zero-Shot Object Detection & Segmentation

SUMMARY YOLOE extends the YOLO family with zero-shot object detection and instance segmentation, letting you specify target classes at inference time rather than training a dedicated model for each task. This tutorial walks through environment setup on a Colab T4 GPU, pulling YOLOE weights from Hugging Face, running text-

7 Aug 2025 • 5 min read

GPT-5 for Vision: Results from 80+ Real-World Tests

SUMMARY Roboflow tested GPT-5 across more than 80 structured visual tasks spanning document understanding, defect detection, object counting, and the RF100-VL object detection benchmark. On Vision Checkup, GPT-5 tied for first place with GPT o4 Mini, driven largely by strong spatial reasoning, but scored only 4 of

7 Aug 2025 • 9 min read

How to Import Supervisely Datasets to Roboflow

Learn how to import datasets from Supervisely into Roboflow, train a YOLOv11 model, and deploy it to the cloud or edge. See how to convert annotations, fine-tune your model, and build production-ready vision workflows.

6 Aug 2025 • 4 min read

Reduce Jittery and Flickering Detections in Computer Vision

SUMMARY Bounding boxes that jump or flicker between frames are a common problem in computer vision pipelines, caused by inconsistent frame-to-frame model detections. The Supervision Detection Smoother blends past bounding box positions using ByteTrack to produce steadier, more consistent detections across frames. The result is cleaner visual output

5 Aug 2025 • 19 min read

Automated Body Measurement with Computer Vision

Learn how to build a real-time body measurement app using computer vision. This guide walks through detecting 17 key body points, and calculating measurements like height, arm span, and shoulder width from a single image.

4 Aug 2025 • 14 min read

Real-Time PCB Defect Detection with Computer Vision

Learn how to build a real-time PCB defect detection system using Roboflow and Python. This step-by-step guide shows how to train a custom model, create an automated inspection workflow, and display results in a sleek, operator-friendly GUI.

4 Aug 2025 • 9 min read

Automate Image Background Blurring Using SAM 2

Learn how to build an automated background blurring workflow using Roboflow and SAM 2, Meta’s state-of-the-art segmentation model. This guide walks through every step.

4 Aug 2025 • 9 min read

How to Create a YOLOv11 Android App

SUMMARY This tutorial walks through training a custom YOLOv11 object detection model with Roboflow and deploying it directly on an Android device to build a real-time coin counting app. The model weights are converted from a .pt file to TorchScript format so they can run on-device without a

1 Aug 2025 • 4 min read

Launch: Verified Model Metrics

SUMMARY Model evaluation scores are not always comparable across providers because teams use different test sets, confidence thresholds, and calculation methods, which makes selecting models for production unreliable. Roboflow Verified Model Metrics addresses this by calculating mAP using open-source COCO Evaluation methodology from the Supervision library on independent test

30 Jul 2025 • 11 min read

Automated Image Tagging with Computer Vision

Learn how to build a Python application that uses computer vision models to analyze images for face visibility, framing, lighting, clutter, and an overall photo score, embedding these tags directly into the image file.

29 Jul 2025 • 18 min read

Model Deployment

In this article you will understand what is ML model deployment, how it works, best practices for model deployment, popular deployment tools and how to deploy a computer vision model on your local machine using Roboflow.

29 Jul 2025 • 8 min read

Real-Time Process Tracking with Computer Vision

With this project, we integrate real-time feedback and computer vision to develop a hand-washing steps-tracking system using a Python application and a Roboflow-trained model.

28 Jul 2025 • 14 min read

AI Data Labeling Guide

Even with advancements in multimodal foundation models, auto-segmentation techniques, and synthetic data generation, accurate AI data labeling remains a crucial component of production-ready AI systems.

28 Jul 2025 • 5 min read

Building a Drowning Detection Model with Computer Vision

Drowning can happen in seconds, even with lifeguards watching. Learn how to build a computer vision system to detect when someone stays underwater too long and trigger life-saving alerts.

25 Jul 2025 • 3 min read

Roboflow Launches 'Roboflow for Startups', Sponsors Vercel AI Accelerator

SUMMARY Roboflow launched the Roboflow for Startups Accelerator Partner Program, which gives early-stage startups accepted into participating accelerators (including Y Combinator, Vercel AI Accelerator, and Sequoia Arc) one free year of the Growth Plan, 1,800 platform credits, priority support, and optional co-marketing to Roboflow's developer

25 Jul 2025 • 9 min read

How to Build an iOS App with Visual AI Capabilities

SUMMARY This tutorial walks through building an iOS app that detects glasses in real time using a custom RF-DETR Nano model trained in Roboflow and deployed via the roboflow-swift SDK. RF-DETR Nano is chosen specifically for its low latency on live camera streams, making it well suited

24 Jul 2025 • 19 min read

What Is Depth Anything V2: Depth Estimation

Understand what depth estimation is. Learn how Depth Anything V2 works and how to use it in Roboflow Workflows.

24 Jul 2025 • 6 min read

Advancing State of the Art Object Detection (Again) with RF-DETR

Learn how to use and train the three new state-of-the-art RF-DETR object detection model sizes.

22 Jul 2025 • 8 min read

Detect NBA 3 Second Violations with AI

SUMMARY Automatically detecting NBA 3-second paint violations requires solving several distinct sub-problems at once: tracking each player across frames, identifying the painted zone, and counting how long each player stays inside it. This tutorial builds that system by combining player tracking with Meta's SAM2, court keypoint

18 Jul 2025 • 5 min read

Use Qwen2.5-VL for Zero-Shot Object Detection

SUMMARY Qwen2.5-VL is an open source vision-language model available in 3B, 7B, and 72B sizes that can locate and label objects in an image from a plain-text prompt, with no labeled training data or fine-tuning required. This tutorial runs the 7B variant on a T4

18 Jul 2025 • 9 min read

How Computer Vision Is Reshaping The Restaurant Industry

Learn how computer vision can be used to improve efficiency in the restaurant industry.

18 Jul 2025 • 7 min read

Use Gemini 2.5 for Zero-Shot Object Detection & Segmentation

Learn how to use Gemini 2.5 for zero-shot object detection and segmentation.

17 Jul 2025 • 6 min read

How to Make a Heatmap with Computer Vision

This guide walks you through building a computer vision-powered heatmap using Roboflow. We'll use the example of mapping my dog’s movement in my apartment - from collecting video data to generating the heatmap by sending data from the workflows directly into a custom Python script.

17 Jul 2025 • 31 min read

AI in Robotics

This blog explores how AI powers modern robotics, from core technologies like computer vision,NLP and ML to real-world robots, models, and projects, revealing how intelligent machines are transforming industry and society.