Forecasting the future trajectories of multiple agents is a core technology for human-robot interaction systems. To predict multi-agent trajectories more accurately, it is inevitable that models need to improve interpretability and reduce redundancy. However, many methods adopt implicit weight calculation or black-box networks to learn the semantic interaction of agents, which obviously lack enough interpretation. In addition, most of the existing works model the relation among all agents in a one-to-one manner, which might lead to irrational trajectory predictions due to its redundancy and noise. To address the above issues, we present Hypertron, a human-understandable and lightweight hypergraph-based multi-agent forecasting framework, to explicitly estimate the motions of multiple agents and generate reasonable trajectories. The framework explicitly interacts among multiple agents and learns their latent intentions by our coarse-to-fine hypergraph convolution interaction module. Our experiments on several challenging real-world trajectory forecasting datasets show that Hypertron outperforms a wide array of state-of-the-art methods while saving over 60% parameters and reducing 30% inference time.

We present a dataset for building detection and classification from very high-resolution satellite imagery with the focus on object-level interpretation of individual buildings. It is meant to provide not only a flexible test platform for object detection algorithms but also a solid basis for the comparison of city morphologies and the investigation of urban planning. In most current open datasets, buildings are treated either as a class of landcover in the form of masks or as simple objects defined by separate contours (footprints). Our dataset, instead, represents individual buildings using in-depth object-level descriptions concerning geometry as well as functionality. Buildings are treated as objects with individual ID and boundary. Adjacent building blocks are also separated according to house numbers making a subsequent high-level classification of individual buildings possible. The buildings are classified into predefined roof types, such as flat, gable and hipped roof as well as functional purposes, i.e., residential, commercial, industrial, public, and their sub-classes, e.g., single-family house, office building and school. In the first version of the dataset we provide selected urban areas from two cities: Beijing in China and Munich in Germany. It, therefore, (1) allows to verify algorithms that are not only valid for specific regions but also work robustly in spite of the diversity of cities on different continents with various land forms and styles of architecture and at the same time (2) provides the possibility to quantitatively compare the statistics and morphology of different cities. It is planned to extend the dataset by a continuous integration of various urban areas worldwide.