English

Generating Contextually-Relevant Navigation Instructions for Blind and Low Vision People

Computation and Language 2024-07-12 v1 Human-Computer Interaction

Abstract

Navigating unfamiliar environments presents significant challenges for blind and low-vision (BLV) individuals. In this work, we construct a dataset of images and goals across different scenarios such as searching through kitchens or navigating outdoors. We then investigate how grounded instruction generation methods can provide contextually-relevant navigational guidance to users in these instances. Through a sighted user study, we demonstrate that large pretrained language models can produce correct and useful instructions perceived as beneficial for BLV users. We also conduct a survey and interview with 4 BLV users and observe useful insights on preferences for different instructions based on the scenario.

Keywords

Cite

@article{arxiv.2407.08219,
  title  = {Generating Contextually-Relevant Navigation Instructions for Blind and Low Vision People},
  author = {Zain Merchant and Abrar Anwar and Emily Wang and Souti Chattopadhyay and Jesse Thomason},
  journal= {arXiv preprint arXiv:2407.08219},
  year   = {2024}
}

Comments

Accepted as RO-MAN 2024 Late Breaking Report

R2 v1 2026-06-28T17:36:48.159Z