TrajNet++ : Interaction-centric Human Trajectory Forecasting Benchmark
Current neural network-based human trajectory forecasting models have been trained on limited data without an explicit focus on socially-interacting trajectories. We present TrajNet++, a large-scale interaction-centric benchmark.
TRANS: Pedestrian Stop and Go
Benchmark for prediction of whether pedestrians will stop walking (Stop) or start to walk (Go) in the near future, for better trajectory prediction around road traffic.
HHI-Assist: A Dataset and Benchmark of Human-Human Interaction in Physical Assistance Scenario
The increasing labor shortage and aging population underscore the need for assistive robots to support human care recipients. Developing such robots requires accurate human motion prediction to ensure their responsiveness and safety. This task is challenging due to the variability in scenarios and the necessity of modeling interactions between agents. To address these challenges, we present HHI-Assist, a collection of motion capture clips of human-human interaction (HHI) for physical assistance.