Representing Spatial Trajectories as Distributions

Didac Suris Coll-Vinent, Carl Vondrick

Keywords: representation learning, human pose, video

Abstract: We introduce a representation learning framework for spatial trajectories. We represent partial observations of trajectories as probability distributions in a learned latent space, which characterize the uncertainty about unobserved parts of the trajectory. Our framework allows us to obtain samples from a trajectory for any continuous point in time—both interpolating and extrapolating. Our flexible approach supports directly modifying specific attributes of a trajectory, such as its pace, as well as combining different partial observations into single representations. Experiments show our method's superiority over baselines in prediction tasks.

TL;DR: We present a representation learning framework for trajectories that models segments of trajectories as distributions in a learned latent space.

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/arxiv:2210.01322/code)