Learning Continuous Environment Fields via Implicit Functions
This work addresses scene representation for robotics and human behavior modeling, offering a novel method for continuous distance encoding, but it is incremental as it builds on neural implicit functions and generative models.
The paper tackles the problem of representing scenes for agent navigation and human trajectory prediction by introducing a continuous environment field that encodes reaching distances to goals along feasible paths, and demonstrates its application in 2D mazes and 3D indoor scenes with efficient and accurate trajectory generation.
We propose a novel scene representation that encodes reaching distance -- the distance between any position in the scene to a goal along a feasible trajectory. We demonstrate that this environment field representation can directly guide the dynamic behaviors of agents in 2D mazes or 3D indoor scenes. Our environment field is a continuous representation and learned via a neural implicit function using discretely sampled training data. We showcase its application for agent navigation in 2D mazes, and human trajectory prediction in 3D indoor environments. To produce physically plausible and natural trajectories for humans, we additionally learn a generative model that predicts regions where humans commonly appear, and enforce the environment field to be defined within such regions. Extensive experiments demonstrate that the proposed method can generate both feasible and plausible trajectories efficiently and accurately.