Addressing Training Bias via Automated Image Annotation

TL;DR

This paper explores an automated image annotation system leveraging wireless localization and camera data to improve training datasets for deep neural networks, highlighting benefits, challenges, and future research directions.

Contribution

It introduces a novel approach combining wireless localization with camera data for automated image annotation, addressing bias in training data for deep learning models.

Findings

Feasibility demonstrated for pedestrian and vehicle detection

Benefits include reduced manual annotation effort

Challenges involve privacy and technical development

Abstract

Build accurate DNN models requires training on large labeled, context specific datasets, especially those matching the target scenario. We believe advances in wireless localization, working in unison with cameras, can produce automated annotation of targets on images and videos captured in the wild. Using pedestrian and vehicle detection as examples, we demonstrate the feasibility, benefits, and challenges of an automatic image annotation system. Our work calls for new technical development on passive localization, mobile data analytics, and error-resilient ML models, as well as design issues in user privacy policies.