A Dataset for Crucial Object Recognition in Blind and Low-Vision Individuals' Navigation

TL;DR

This paper presents a new dataset of videos and object taxonomy tailored for improving real-time object recognition systems to assist blind and low-vision individuals in navigation, highlighting gaps in existing datasets and the need for specialized resources.

Contribution

The paper introduces a novel dataset with 21 videos and a taxonomy of 90 crucial objects for BLV navigation, addressing limitations of existing datasets and supporting development of inclusive navigation aids.

Findings

Most existing datasets lack key objects relevant to BLV navigation.

State-of-the-art models perform poorly on the new dataset for critical objects.

The dataset is publicly available for further research.

Abstract

This paper introduces a dataset for improving real-time object recognition systems to aid blind and low-vision (BLV) individuals in navigation tasks. The dataset comprises 21 videos of BLV individuals navigating outdoor spaces, and a taxonomy of 90 objects crucial for BLV navigation, refined through a focus group study. We also provide object labeling for the 90 objects across 31 video segments created from the 21 videos. A deeper analysis reveals that most contemporary datasets used in training computer vision models contain only a small subset of the taxonomy in our dataset. Preliminary evaluation of state-of-the-art computer vision models on our dataset highlights shortcomings in accurately detecting key objects relevant to BLV navigation, emphasizing the need for specialized datasets. We make our dataset publicly available, offering valuable resources for developing more inclusive navigation systems for BLV individuals.