Latest Posts

Scalable Polygon Annotation: Clean, Aligned Boundaries for Geospatial Data
Stan Soldatov
•

Scalable Polygon Annotation: Clean, Aligned Boundaries for Geospatial Data

Annotate buildings, parcels and land cover with polygons in Supervisely: start from AI masks, align boundaries between objects, cut holes and fix local errors.

Introducing Video + GPS Telemetry: Synchronize Footage, Maps, and Sensor Data
Stan Soldatov
•

Introducing Video + GPS Telemetry: Synchronize Footage, Maps, and Sensor Data

Review video with synchronized GPS maps and sensor charts in Supervisely, find events by route or telemetry, and label real-world footage for Physical AI.

Ultimate Guide On Multiview Image Annotation For Custom Training Datasets in Computer Vision
Almaz Gimaev
•

Ultimate Guide On Multiview Image Annotation For Custom Training Datasets in Computer Vision

Learn how to use Supervisely's Multiview Image Labeling Toolbox to upload and annotate multiview images simultaneously.

Introducing Video Annotation 3.0
Denis Drozdov
•

Introducing Video Annotation 3.0

Our largest update so far, with AI-powered auto tracking, enhanced tag management, and a sleek new design for faster, smarter video labeling.

Introducing MLOps Workflow and Data Versioning: track your datasets, models and Computer Vision experiments
Sergey Sych
•

Introducing MLOps Workflow and Data Versioning: track your datasets, models and Computer Vision experiments

Easy-to-use visual MLOps Pipelines simplify data version control and the ML modeling process, making all experiments reproducible and clear.

Introducing Definitions Panel: a new way to assign classes and tags
Denis Drozdov
•

Introducing Definitions Panel: a new way to assign classes and tags

How to reduce the number of steps and clicks to label an object or change its class.

Segment Anything 2 (SAM2) in Supervisely: The Fast and Accurate Object Segmentation Tool for Image and Video Labeling
Max Teselkin
•

Segment Anything 2 (SAM2) in Supervisely: The Fast and Accurate Object Segmentation Tool for Image and Video Labeling

How to automatically segment and track objects on images and videos using Segment Anything model 2 in Supervisely

Complete guide on 2D medical image annotation in Supervisely
Ars Vladimirov
•

Complete guide on 2D medical image annotation in Supervisely

Learn how to annotate DICOM, NRRD, NIfTI medical images in 2D with efficient collaboration, CT / MRI windowing and AI auto-labeling.

The Complete Guide to Image Matting: Segment using Matte Anything Model and Brush Tool with Transparency
Almaz Gimaev
•

The Complete Guide to Image Matting: Segment using Matte Anything Model and Brush Tool with Transparency

Learn how to annotate images for image matting task in Supervisely using predictions from Matte Anything model.

How to use 3D Slicer with Supervisely to Annotate Medical Training Datasets At Scale | Computer Vision Guide
Sergey Sych
•

How to use 3D Slicer with Supervisely to Annotate Medical Training Datasets At Scale | Computer Vision Guide

Discover how to unlock the power of annotating 3D medical images and boost teamwork collaboration by using the best of both worlds.

Top 10 Quality Assurance Tools And Interactive Statistics For Your Training Dataset | Ultimate Guide
Lisa Uspenyeva
•

Top 10 Quality Assurance Tools And Interactive Statistics For Your Training Dataset | Ultimate Guide

Learn how to use best quality assurance and interactive statistical tools to perfect your custom training datasets and improve neural network performance.

Introducing Advanced Dataset Filters for Efficient Data Management | Supervisely Tutorial
Lisa Uspenyeva
•

Introducing Advanced Dataset Filters for Efficient Data Management | Supervisely Tutorial

Learn how to use conditional filters and build custom queries to your training data. Easily search, filter, sample and explore your Computer Vision datasets.

The Ultimate Image Annotation Tool: The Complete Guide to New Supervisely Labeling Toolbox
Lisa Uspenyeva
•

The Ultimate Image Annotation Tool: The Complete Guide to New Supervisely Labeling Toolbox

Learn how to use new Supervisely Image Annotation Tool designed for fast and accurate image annotation in Computer Vision.

Manual Image Segmentation in Computer Vision: A Comprehensive Overview of Annotation Techniques
Lisa Uspenyeva
•

Manual Image Segmentation in Computer Vision: A Comprehensive Overview of Annotation Techniques

An overview of the three main manual image segmentation methods in computer vision: overlaying, snapping to object boundaries and splitting masks.

Sidebar is back! New layout and more — February 2024 updates
Denis Drozdov
•

Sidebar is back! New layout and more — February 2024 updates

Learn about the new sidebar that offers quick access to frequently used pages, recent apps and updated team and workspace context switchers.

How To Segment Entire Object And Its Component Parts With Mask Splitting Method
Lisa Uspenyeva
•

How To Segment Entire Object And Its Component Parts With Mask Splitting Method

Complete tutorial on labeling multiple parts of the object using efficient splitting annotation technique in Computer Vision

How To Tag Multi-Camera Video Segments For Action Recognition In Computer Vision | Full Tutorial
Lisa Uspenyeva
•

How To Tag Multi-Camera Video Segments For Action Recognition In Computer Vision | Full Tutorial

Guide on video annotation, tagging and management tool tailored for multi-camera videos in Supervisely

Guide to Training Custom Interactive Instance Segmentation Model for Agricultural Images
Grigoriy Mananian
•

Guide to Training Custom Interactive Instance Segmentation Model for Agricultural Images

We trained Instance Segmentation model for agricultural plant images and achieved remarkable results.

Image and object tags for classification in Computer Vision: Complete Guide
Lisa Uspenyeva
•

Image and object tags for classification in Computer Vision: Complete Guide

Tutorial on how to use tags and attribute of different types in various industries with real examples.

How to Annotate Multispectral Images for Computer Vision Models
Stan Soldatov
•

How to Annotate Multispectral Images for Computer Vision Models

Learn how to use multi-view display in Supervisely Image Labeling Tool to efficiently annotate multispectral images

How To Use Model Assisted Labeling To Boost Annotation Performance With Supervisely
Lisa Uspenyeva
•

How To Use Model Assisted Labeling To Boost Annotation Performance With Supervisely

A Deep Dive into the Beloved Supervisely Smart Labeling Tools for AI-Assisted Interactive Object Segmentation

Brush Annotation Tool: your must-have for Object Segmentation
Lisa Uspenyeva
•

Brush Annotation Tool: your must-have for Object Segmentation

Learn how to annotate objects of any complexity by creating freeform outlines using the Brush annotation tool.

Mask Pen Annotation Tool for Object Segmentation in Supervisely
Lisa Uspenyeva
•

Mask Pen Annotation Tool for Object Segmentation in Supervisely

Learn how to use Mask Pen annotation tool as a combination of polygonal contours and free-form drawing contours.

How to Train a Model with Only 62 Labeled Images using Semi-Supervised Learning
Max Eliseev
•

How to Train a Model with Only 62 Labeled Images using Semi-Supervised Learning

This experiment reveals the potential of semi-supervised learning, proving the opportunity to train a model using only a small fraction of labeled data.

Polygon Annotation Best Practices for Semantic & Instance Segmentation
Lisa Uspenyeva
•

Polygon Annotation Best Practices for Semantic & Instance Segmentation

Learn how to use Polygon tool for efficient and precise object segmentation in Computer Vision.

Best Bounding Box Image Annotation Tools For Object Detection - Complete Overview
Lisa Uspenyeva
•

Best Bounding Box Image Annotation Tools For Object Detection - Complete Overview

How to use modern bounding box image labeling toolbox for Computer Vision in Supervisely.

How To Use 3D Object Interpolation To Speed Up Point Cloud Annotation for LiDAR & Radar
Max Teselkin
•

How To Use 3D Object Interpolation To Speed Up Point Cloud Annotation for LiDAR & Radar

The complete guide on 3D bounding box interpolation in point cloud episodes in Supervisely.

What is Sliding Window in Object Detection: Complete Overview of Methods & Tools
Sergey Sych
•

What is Sliding Window in Object Detection: Complete Overview of Methods & Tools

How to use sliding window method with object detection models to improve accuracy in Computer Vision.

Unleash The Power of Domain Adaptation - How to Train Perfect Segmentation Model on Synthetic Data with HRDA
Max Eliseev
•

Unleash The Power of Domain Adaptation - How to Train Perfect Segmentation Model on Synthetic Data with HRDA

Tutorial on training accurate and robust Semantic Segmentation model without manual annotation using only synthetic training dataset

Labeling Queues: Streamline Your Labeling Pipeline
Lisa Uspenyeva
•

Labeling Queues: Streamline Your Labeling Pipeline

Discover the optimal approach for annotating and structuring large-scale data labeling tasks with using Labeling Queues and following our step-by-step guide.

Lessons Learned From Training a Segmentation Model On Synthetic Data
Max Eliseev
•

Lessons Learned From Training a Segmentation Model On Synthetic Data

Insights from training dozens of segmentation models on synthetic data, and methods that improve the quality of synthetic training data.

Mastering Labeling Jobs: Your Ultimate Guide
Lisa Uspenyeva
•

Mastering Labeling Jobs: Your Ultimate Guide

Discover the power of collaborative annotation through labeling jobs. Learn how to harness teamwork for efficient and accurate data annotation.

Dataset Ninja 🄷 — the best way to Search and Explore Computer Vision Datasets
Max Kolomeychenko
•

Dataset Ninja 🄷 — the best way to Search and Explore Computer Vision Datasets

High-quality training datasets with deep analysis and visualization tools.

Human Pose Estimation: all you need to know
Lisa Uspenyeva
•

Human Pose Estimation: all you need to know

Get ready to be impressed by how quickly and accurately Supervisely can estimate human poses using advanced techniques.

Unlocking Text -  OCR with Supervisely
Valeria Vorozhko
•

Unlocking Text - OCR with Supervisely

Discover how to make the most of Optical Character Recognition (OCR) capabilities within Supervisely.

The Complete Guide to Animal Pose Estimation: Tools, Models, Tutorial
Lisa Uspenyeva
•

The Complete Guide to Animal Pose Estimation: Tools, Models, Tutorial

Discovering animal pose estimation with Supervisely. Learn how to annotate animals using keypoints and State-of-the-Art neural networks.

ViTPose — How to use the best Pose Estimation Model on Humans & Animals
Max Kolomeychenko
•

ViTPose — How to use the best Pose Estimation Model on Humans & Animals

The complete guide on automatic body pose estimation of animals and humans on your images in Supervisely.

Introducing Supervisely Synthetic Crack Segmentation Dataset
Max Eliseev
•

Introducing Supervisely Synthetic Crack Segmentation Dataset

We present our synthetic dataset for road surface crack segmentation that was generated automatically, available for research purposes.

How to Train Smart Tool for Precise Cracks Segmentation in Industrial Inspection
Valeria Vorozhko
•

How to Train Smart Tool for Precise Cracks Segmentation in Industrial Inspection

Step-by-step guide for industrial inspection cracks segmentation on images using custom interactive AI model.

How to run OpenAI CLIP with UI for Image Retrieval and Filtering your dataset
Max Kolomeychenko
•

How to run OpenAI CLIP with UI for Image Retrieval and Filtering your dataset

How to use text prompts and AI to search and query relevant images in your training datasets.

Segment Anything in High Quality (HQ-SAM): a new Foundation Model for Image Segmentation (Tutorial)
Max Kolomeychenko
•

Segment Anything in High Quality (HQ-SAM): a new Foundation Model for Image Segmentation (Tutorial)

How to use new version of Segment Anything + detailed comparison with original SAM model.

Complete Guide to Object Tracking: Best AI Models, Tools and Methods
Max Kolomeychenko
•

Complete Guide to Object Tracking: Best AI Models, Tools and Methods

In this ultimate guide and tutorial you will learn what is object tracking and learn how to track objects on your videos with the best models and tools.

XMem + Segment Anything: Video Object Segmentation SOTA | Tutorial
Max Kolomeychenko
•

XMem + Segment Anything: Video Object Segmentation SOTA | Tutorial

How to segment and track objects on videos automatically with Segment Anything and XMem to build custom training datasets.

Automate manual labeling with custom interactive segmentation model for agricultural images
Valeria Vorozhko
•

Automate manual labeling with custom interactive segmentation model for agricultural images

How to speed up image segmentation in agriculture with custom AI models

How to auto-label images with OWL-ViT - SOTA Google's foundation object detector
Max Kolomeychenko
•

How to auto-label images with OWL-ViT - SOTA Google's foundation object detector

Automatically detect anything with only one example using OWL-ViT - one-shot object detection SOTA

No-code tutorial: train and predict YOLOv8 on custom data
Max Kolomeychenko
•

No-code tutorial: train and predict YOLOv8 on custom data

The easiest way to get custom YOLOv8 model trained on your own dataset and deploy it with zero coding in the browser.

Best DICOM & NIfTI annotation tools for Medical Imaging AI
Sergey Sych
•

Best DICOM & NIfTI annotation tools for Medical Imaging AI

Learn how to label CT, MRI, and PET medical images for your computer vision models.

Top 5 AI tools for fast surgical video annotation
Lisa Uspenyeva
•

Top 5 AI tools for fast surgical video annotation

How to use modern AI object tracking and segmentation tools to get medical training data 10x faster

How to use labeling consensus to get accurate training data
Max Kolomeychenko
•

How to use labeling consensus to get accurate training data

Consensus with detailed reports allows you to monitor and prevent systematic inconsistencies during the labeling process

Develop AI faster and better

Thousands of companies and researchers use Supervisely every day to label assets, manage datasets, collaborate together and train neural networks.