Convolutions for Spatial Interaction Modeling
This work provides an efficient alternative for time-critical systems like autonomous vehicles, though it is incremental as it revisits existing methods.
The paper tackled the problem of modeling spatial interactions for predicting actor motion around autonomous vehicles, showing that 2D convolutions achieve comparable performance to graph neural networks with lower latency, and introduced a novel interaction loss to enhance modeling.
In many different fields interactions between objects play a critical role in determining their behavior. Graph neural networks (GNNs) have emerged as a powerful tool for modeling interactions, although often at the cost of adding considerable complexity and latency. In this paper, we consider the problem of spatial interaction modeling in the context of predicting the motion of actors around autonomous vehicles, and investigate alternatives to GNNs. We revisit 2D convolutions and show that they can demonstrate comparable performance to graph networks in modeling spatial interactions with lower latency, thus providing an effective and efficient alternative in time-critical systems. Moreover, we propose a novel interaction loss to further improve the interaction modeling of the considered methods.