DAG-Net: Double Attentive Graph Neural Network for Trajectory Forecasting

26 May 2020  ·  Alessio Monti, Alessia Bertugli, Simone Calderara, Rita Cucchiara ·

Understanding human motion behaviour is a critical task for several possible applications like self-driving cars or social robots, and in general for all those settings where an autonomous agent has to navigate inside a human-centric environment. This is non-trivial because human motion is inherently multi-modal: given a history of human motion paths, there are many plausible ways by which people could move in the future... Additionally, people activities are often driven by goals, e.g. reaching particular locations or interacting with the environment. We address the aforementioned aspects by proposing a new recurrent generative model that considers both single agents' future goals and interactions between different agents. The model exploits a double attention-based graph neural network to collect information about the mutual influences among different agents and to integrate it with data about agents' possible future objectives. Our proposal is general enough to be applied to different scenarios: the model achieves state-of-the-art results in both urban environments and also in sports applications. read more

PDF Abstract


  Add Datasets introduced or used in this paper

Results from the Paper

 Ranked #1 on Trajectory Prediction on Stanford Drone (ADE (in world coordinates) metric)

     Get a GitHub badge
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Trajectory Prediction Stanford Drone DAG-Net ADE (in world coordinates) 0.54 # 1
FDE (in world coordinates) 1.05 # 2
Trajectory Prediction STATS SportVu NBA [ATK] DAG-Net ADE 9.18 # 1
FDE 13.54 # 1
Trajectory Prediction STATS SportVu NBA [DEF] DAG-Net ADE 7.01 # 1
FDE 9.76 # 1


No methods listed for this paper. Add relevant methods here