Skip to content

Vidat Overview

Vidat is a browser-based video annotation platform developed by the Computer Vision and Machine Learning (CVML) Group at the Australian National University (ANU).
Designed for efficiency and ease of use, Vidat enables precise, frame-level annotation and tracking for a wide range of computer vision applications.

Key Features

  • Comprehensive Video Annotation
    Create detailed annotations by drawing bounding boxes, regions, and skeletons, labeling multiple objects, and tracking them seamlessly across frames.

  • Intuitive Web Interface
    Vidat requires no installation. Simply open the platform in your browser, upload your video, and start annotating immediately.

  • Standardized Export
    Export annotations in JSON format, compatible with common computer vision workflows and machine learning pipelines.

  • Open-Source and Research-Driven
    Developed and maintained by the ANU CVML Group, Vidat is open-source and continually improved for research and practical deployment.

Vidat supports the entire lifecycle of video annotation — from frame-level object detection to tracking and export — within an accessible, browser-based environment.
It empowers researchers, developers, and practitioners to efficiently create high-quality labeled datasets for computer vision and machine learning projects.

Installation

Vidat is a web-based video annotation tool that runs directly in your browser.
It does not require local installation for standard use. However, it can also be installed and self-hosted for private or offline deployment.

Quick Start (No Installation Required)

To use Vidat immediately:

  1. Visit the Vidat Tool
  2. Ensure your browser supports modern JavaScript features
  3. No login or setup required — simply upload a video and begin annotating

When using Vidat from the official ANU or Aliyun hosts, your video data never leaves your local machine. All annotation operations occur entirely in your browser. No video data is uploaded to any server.

Self-Hosted Installation (Optional)

Vidat can also be hosted on your own computer or private server.
This option is recommended if you require:

  • Integration with internal data pipelines or APIs
  • A private or offline deployment