Combining lightweight object detection with Large Language Models enhances safety in visual navigation for the visually impaired.
Large Language Models (LLMs) enhance real-time anomaly detection for safe visual navigation.