WearVQA: A Visual Question Answering Benchmark for Wearables in Egocentric Authentic Real-world scenarios

TL;DR

WearVQA is a new benchmark designed to evaluate the visual question answering capabilities of AI models on wearable devices in real-world, egocentric scenarios with diverse challenges and image qualities.

Contribution

This paper introduces WearVQA, the first benchmark focusing on egocentric, real-world wearable scenarios, with a comprehensive dataset and evaluation framework for VQA tasks.

Findings

Open-source models achieve 24-52% accuracy on WearVQA.

Performance drops significantly on low-quality images and reasoning tasks.

WearVQA presents a challenging benchmark for advancing wearable AI systems.

Abstract

We introduce WearVQA, the first benchmark specifically designed to evaluate the Visual Question Answering (VQA) capabilities of multi-model AI assistant on wearable devices like smart glasses. Unlike prior benchmarks that focus on high-quality, third-person imagery, WearVQA reflects the unique challenges of ego-centric interaction-where visual inputs may be occluded, poorly lit, unzoomed, or blurry, and questions are grounded in realistic wearable use cases. The benchmark comprises 2,520 carefully curated image-question-answer triplets, spanning 7 diverse image domains including both text-centric and general scenes, 10 cognitive task types ranging from basic recognition to various forms of reasoning, and 6 common wearables-specific image quality issues. All questions are designed to be answerable using only the visual input and common senses. WearVQA is paired with a rigorous LLM-as-a-judge evaluation framework with 96% labeling accuracy. Open-source and proprietary multi-model LLMs achieved a QA accuracy as low as 24-52% on WearVQA, with substantial drops on lower-quality images and reasoning-heavy tasks. These observations position WearVQA as a comprehensive and challenging benchmark for guiding technical advancement towards robust, real-world multi-model wearables AI systems.