Automatic image tagging using Data Explorer

Introduction

A dataset of images, used for computer vision tasks, could be the key to success or failure. A clean dataset could lead the way to a great algorithm, model and ultimately system, while no matter how good the model or algorithm is, garbage in — garbage out.

Data annotation, especially in the vision world, is a known and much needed task, with many limitations and very few fully automated solutions.

Automatic tagging

In order to address the annotation problems, Akridata brought the latest in automatic tagging into?Data Explorer. Through its Pipelines, a dataset can be automatically tagged, using any foundation, or standard, model.

As an example, a pipeline based on the?RAM foundation model?was built and applied to the?nuScenes?dataset.

The dataset was processed and a list of tags was generated per image — as seen in the example below:

After using Data Explorer to?visualize?a dataset, automatic tagging as described in this blog, could be used for classification problems, assist with further example?searching?and other related tasks.

Summary

In this blog we saw how to automatically tag a dataset of images using Data Explorer. This could be used for classification problems, assisting with searching through the dataset and much more.

Message me at [email protected] for a demo of Data Explorer, visit us at?akridata.ai?or click here to?register?for a free account.