HEADS-UP is the first egocentric dataset collected from head-mounted cameras, designed specifically for trajectory prediction in blind assistance systems. Given the large population of blind and visually impaired individuals, there is an increasing need for intelligent assistive tools that can provide real-time warnings about potential collisions with dynamic obstacles. These systems depend on algorithms that can predict the trajectories of moving objects, such as pedestrians, to issue timely hazard alerts. However, current datasets do not capture the essential information from the perspective of a blind individual. To bridge this gap, we introduce HEADS-UP, a novel dataset specifically designed for trajectory prediction in this context.
