Track Anything Annotate: Video annotation and dataset generation of computer vision models

TL;DR

This paper introduces Track Anything Annotate, a tool that leverages video tracking and segmentation to rapidly generate annotated datasets, significantly reducing manual effort in dataset creation for computer vision models.

Contribution

The paper presents a novel prototype that automates video annotation and dataset generation, improving efficiency over traditional manual methods.

Findings

Prototype accelerates dataset creation process

Significant reduction in manual annotation effort

Resources available for public use

Abstract

Modern machine learning methods require significant amounts of labelled data, making the preparation process time-consuming and resource-intensive. In this paper, we propose to consider the process of prototyping a tool for annotating and generating training datasets based on video tracking and segmentation. We examine different approaches to solving this problem, from technology selection through to final implementation. The developed prototype significantly accelerates dataset generation compared to manual annotation. All resources are available at https://github.com/lnikioffic/track-anything-annotate