Multimodal Future Localization and Emergence Prediction for Objects in Egocentric View with a Reachability Prior
This addresses safety and navigation challenges for autonomous vehicles by enhancing anticipation of dynamic objects without relying on structural maps, though it is incremental in combining existing techniques.
The paper tackles the problem of predicting future locations and emergence of objects in egocentric vehicle views, using a reachability prior from semantic maps and multi-hypotheses learning, resulting in improved multimodal predictions and zero-shot transfer to unseen datasets.
In this paper, we investigate the problem of anticipating future dynamics, particularly the future location of other vehicles and pedestrians, in the view of a moving vehicle. We approach two fundamental challenges: (1) the partial visibility due to the egocentric view with a single RGB camera and considerable field-of-view change due to the egomotion of the vehicle; (2) the multimodality of the distribution of future states. In contrast to many previous works, we do not assume structural knowledge from maps. We rather estimate a reachability prior for certain classes of objects from the semantic map of the present image and propagate it into the future using the planned egomotion. Experiments show that the reachability prior combined with multi-hypotheses learning improves multimodal prediction of the future location of tracked objects and, for the first time, the emergence of new objects. We also demonstrate promising zero-shot transfer to unseen datasets. Source code is available at $\href{https://github.com/lmb-freiburg/FLN-EPN-RPN}{\text{this https URL.}}$