Abstract
The increasing popularity of egocentric cameras has generated growing interest in studying multi-camera interactions in shared environments. Although large-scale datasets such as Ego4D and Ego-Exo4D have propelled egocentric vision research, interactions between multiple camera wearers remain underexplored-a key gap for applications like immersive learning and collaborative robotics. To bridge this, we present TF2025, an expanded dataset with synchronized first- and third-person views. In addition, we introduce a sequence-based method to identify first-person wearers in third-person footage, combining motion cues and person re-identification.
Abstract (translated)
第一人称相机日益普及，引发了人们对共享环境中多相机交互研究的兴趣不断增加。尽管大规模数据集如Ego4D和Ego-Exo4D推动了第一人称视觉研究的发展，但多个佩戴者之间的互动仍被严重忽视——这对沉浸式学习和协作机器人等应用来说是一个关键缺口。为了弥补这一不足，我们推出了TF2025，这是一个扩展的数据集，包含了同步的第一人称和第三人称视角视图。此外，我们还介绍了一种基于序列的方法来识别第三人称视频中佩戴第一人称相机的人，该方法结合了运动线索和个人重识别技术。
URL
https://arxiv.org/abs/2506.00394
