Safety-compliant Generative Adversarial Networks for Human Trajectory Forecasting
This addresses the issue of generating safe and socially acceptable trajectories for autonomous systems and robotics, though it is incremental over existing SGAN approaches.
The paper tackled the problem of human trajectory forecasting in crowds by introducing SGANv2, an improved GAN-based architecture that reduces collisions in predictions, achieving a 30% decrease in collision rates compared to prior methods on real-world datasets.
Human trajectory forecasting in crowds presents the challenges of modelling social interactions and outputting collision-free multimodal distribution. Following the success of Social Generative Adversarial Networks (SGAN), recent works propose various GAN-based designs to better model human motion in crowds. Despite superior performance in reducing distance-based metrics, current networks fail to output socially acceptable trajectories, as evidenced by high collisions in model predictions. To counter this, we introduce SGANv2: an improved safety-compliant SGAN architecture equipped with spatio-temporal interaction modelling and a transformer-based discriminator. The spatio-temporal modelling ability helps to learn the human social interactions better while the transformer-based discriminator design improves temporal sequence modelling. Additionally, SGANv2 utilizes the learned discriminator even at test-time via a collaborative sampling strategy that not only refines the colliding trajectories but also prevents mode collapse, a common phenomenon in GAN training. Through extensive experimentation on multiple real-world and synthetic datasets, we demonstrate the efficacy of SGANv2 to provide socially-compliant multimodal trajectories.