GD-GAN: Generative Adversarial Networks for Trajectory Prediction and Group Detection in Crowds
This addresses surveillance and crowd analysis by automating group detection without labeled data, though it appears incremental as it builds on existing GAN approaches.
The paper tackles human trajectory prediction and social group detection in crowds using a generative adversarial network framework that preserves spatio-temporal neighborhood structure, achieving better anticipation of human sociological behavior than state-of-the-art methods on multiple public benchmarks.
This paper presents a novel deep learning framework for human trajectory prediction and detecting social group membership in crowds. We introduce a generative adversarial pipeline which preserves the spatio-temporal structure of the pedestrian's neighbourhood, enabling us to extract relevant attributes describing their social identity. We formulate the group detection task as an unsupervised learning problem, obviating the need for supervised learning of group memberships via hand labeled databases, allowing us to directly employ the proposed framework in different surveillance settings. We evaluate the proposed trajectory prediction and group detection frameworks on multiple public benchmarks, and for both tasks the proposed method demonstrates its capability to better anticipate human sociological behaviour compared to the existing state-of-the-art methods.