Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces ForeSightGuide, an anticipatory framework that enhances electronic travel aids for the visually impaired by integrating semantic scene understanding with predictive hazard assessment. By leveraging the reasoning capabilities of Vision-Language Models (VLMs), ForeSightGuide effectively filters out non-threatening objects, reducing cognitive overload and providing concise guidance in dynamic environments. Experimental results demonstrate its superiority over existing systems, achieving a remarkable reduction in redundant alerts while maintaining a low missed-hazard rate, thereby improving the safety and usability of mobility aids.
ForeSightGuide cuts down redundant alerts to just 0.299 per guidance output, revolutionizing how visually impaired individuals navigate complex environments.
Electronic travel aids are pivotal for the independent mobility of the visually impaired. While Vision-Language Models (VLMs) offer rich environmental understanding, they often suffer from excessive false positives in dynamic scenarios, leading to cognitive overload. To address this, we present ForeSightGuide, an anticipatory assistive guidance framework that couples semantic scene understanding with predictive hazard assessment. Unlike reactive systems, ForeSightGuide leverages the reasoning capabilities of VLMs to anticipate obstacle motion, effectively filtering out non-threatening objects to provide concise, actionable guidance. To validate our approach, we introduce a novel dataset captured in complex, dynamic real-world traffic scenes, designed to benchmark predictive capabilities. Extensive experiments on both public benchmarks and our proposed dataset demonstrate that ForeSightGuide achieves state-of-the-art performance. Notably, it significantly mitigates information overload by reducing redundant alerts to 0.299 per guidance output while maintaining a low missed-hazard rate of 0.112, proving its efficacy for safe walking assistance.