Skip to content
Stork.AIstork.ai
AI Tool

VGG Image Annotator (VIA) Review

VGG Image Annotator (VIA) is a suite of web-based tools for annotating image, audio, and video files, operating offline from single HTML files without requiring a backend server or installation.

shipped Jul 31, 2026image-generationfree
Domain rating91Monthly visits7/mo
image-generationresearch
VGG Image Annotator (VIA) — product screenshot

Why it matters

1Developed by the Visual Geometry Group (VGG) at the University of Oxford.
2Operates entirely offline from a single HTML file, typically under 300-400 KB.
3Supports various annotation shapes including rectangles, polygons, points, and lines.
4Version 3.0.11, released October 8, 2021, added audio/video annotation and magnifier features.

Specs

API Docs

View Documentation →

GitHub

View Repository →

API Available

Yes, public API

overview

What is VGG Image Annotator (VIA)?

VGG Image Annotator (VIA) is an image, audio, and video annotation tool developed by the Visual Geometry Group (VGG) at the University of Oxford that enables researchers and individuals to manually define and describe regions in media files. It operates as a self-contained, web-based application from a single HTML file, requiring no installation or backend server.

features

Key Features of VGG Image Annotator (VIA)

VGG Image Annotator (VIA) provides a set of features designed for manual annotation tasks, emphasizing simplicity and offline functionality. Its core capabilities revolve around versatile region definition and data export.

  • Self-contained HTML file: The entire application runs from a single HTML file, typically under 300-400 KB, requiring no installation.
  • Offline functionality: Operates entirely within the browser, ensuring data privacy as files do not leave the user's machine.
  • Multi-media support: Annotates image, audio, and video files.
  • Diverse annotation shapes: Supports rectangles, polygons, points, lines, ellipses, and circles for region definition.
  • Attribute management: Allows users to assign attributes (key-value pairs) to regions and files.
  • Keyboard shortcuts: Provides shortcuts for efficient annotation, including toggling boundary and label visibility.
  • Magnifier and zoom: Includes features for detailed annotation in the image annotator (added in VIA 3.0.11).
  • Video tracking assistance (VVT): Aids in annotating bounding box video tracks.
  • Python package (python-via): Facilitates handling VIA projects and converting between VIA formats and COCO.
  • Shared/collaborative projects (VIA3, LISA): Supports collaborative annotation workflows.

use cases

Who Should Use VGG Image Annotator (VIA)?

VGG Image Annotator (VIA) is primarily suited for users requiring a straightforward, local, and free annotation solution for computer vision and multimedia tasks. Its design caters to specific project types and user profiles.

  • Academic Researchers: Ideal for creating datasets for computer vision research, object detection, image segmentation, and classification tasks.
  • Small, Solo Projects: Suitable for individual developers or researchers working on projects that do not require large-scale team collaboration or AI-assisted labeling.
  • Beginners in Computer Vision: Provides an accessible entry point for learning manual data annotation without complex setup.
  • Privacy-Sensitive Projects: Beneficial for annotating sensitive data, as all operations occur locally in the browser, ensuring data never leaves the user's machine.
  • Budget-Conscious Users: As a free and open-source tool, it eliminates software costs for annotation needs.

how to use

How to Use VGG Image Annotator (VIA)

VGG Image Annotator (VIA) is designed for immediate use by simply opening its HTML file in a web browser. The process involves loading media, defining regions, and exporting annotations.

  • 1Download the latest VIA HTML file (e.g., via_3.0.11.html) from the official VGG website.
  • 2Open the downloaded HTML file in any modern web browser (e.g., Chrome, Firefox, Edge).
  • 3Load images, audio, or video files into the annotator using the 'Add Files' option.
  • 4Select an annotation tool (e.g., rectangle, polygon, point) and draw regions on the loaded media.
  • 5Assign attributes (e.g., class labels, descriptions) to the annotated regions.
  • 6Export the annotations in JSON format, which can then be converted to other formats like COCO using the python-via package.

pricing

VGG Image Annotator (VIA) Pricing & Plans

VGG Image Annotator (VIA) is distributed as free and open-source software. It operates under the BSD-2 clause license, which permits its use without cost in both academic and commercial environments. There are no paid tiers, subscription models, or usage fees associated with VIA.

  • VGG Image Annotator (VIA): Free

Pros

  • +Free and open-source under the BSD-2 clause license.
  • +Operates entirely offline from a single HTML file, requiring no installation or backend server.
  • +Ensures data privacy as media files never leave the user's local machine.
  • +Supports a wide range of annotation shapes for images, audio, and video.
  • +Simple, lightweight, and easy to use for manual annotation tasks.

Cons

  • −Lacks AI-assisted labeling features, making it time-consuming for large datasets.
  • −Does not offer robust collaboration tools or project management features for teams.
  • −Less scalable for enterprise-level projects compared to commercial annotation platforms.
  • −User interface is minimalist, potentially lacking advanced functionalities found in more complex tools.
  • −Manual nature can be slow for extensive annotation tasks.

Similar Tools

VGG Image Annotator (VIA) vs Competitors

VGG Image Annotator (VIA) occupies a niche as a lightweight, offline, and free manual annotation tool. Its competitive positioning is defined by its simplicity and self-contained nature, contrasting with more feature-rich, often cloud-based, or installation-dependent alternatives.

1
LabelMe↗

A popular Python-based desktop application for image annotation, supporting various shapes like polygons, rectangles, circles, lines, and points.

Unlike VIA's self-contained HTML file that runs directly in a browser without any setup, LabelMe requires a Python environment and installation to run as a desktop application.

2
makesense.aiOn Stork

A user-friendly, browser-based tool that runs entirely client-side for object detection and segmentation tasks, supporting bounding boxes, polygons, points, and lines.

While it runs in the browser, makesense.ai is typically accessed online; VIA is specifically designed for offline use from a single downloaded HTML file, offering greater portability without an internet connection after initial download.

3
ImgLab↗

A lightweight, browser-based image annotation tool that can be run locally from a downloaded HTML file, supporting bounding box, polygon, and point annotations.

ImgLab offers a very similar browser-based, local annotation experience to VIA, but may have a different user interface or specific feature set for annotation tasks.

4
Sloth↗

A highly configurable desktop tool for annotating images and videos, built with Python and PyQt, supporting rectangles, polygons, and points.

As a desktop application, Sloth requires Python and PyQt installation, which is a more involved setup compared to VIA's single HTML file that runs directly in a web browser without any prior installation.

See every VGG Image Annotator (VIA) alternative, compared→

More on Stork

Related AI Tools

Other tools in this category, matched by shared tags

Realword AIIO InventoryNegoti8SnapBlockDraftlyPinokio
Visit VGG Image Annotator (VIA)↗
Connect
⌘
GitHubgithub.com/cdli-gh/coco-convertor/