RED: A Simple but Effective Baseline Predictor for the TrajNet Benchmark

Abstract

In recent years, there is a shift from modeling the tracking problem based on Bayesian formulation towards using deep neural networks. Towards this end, in this paper the effectiveness of various deep neural networks for predicting future pedestrian paths are evaluated. The analyzed deep networks solely rely, like in the traditional approaches, on observed tracklets without human-human interaction information. The evaluation is done on the publicly available TrajNet benchmark dataset [39], which builds up a repository of considerable and popular datasets for trajectory prediction. We show how a Recurrent-Encoder with a Dense layer stacked on top, referred to as RED-predictor, is able to achieve toprank at the TrajNet 2018 challenge compared to elaborated models. Further, we investigate failure cases and give explanations for observed phenomena, and give some recommendations for overcoming demonstrated shortcomings.

What the Constant Velocity Model Can Teach Us About Pedestrian Motion Prediction

“…While the benefits of data augmentation have been reported earlier [2], [14], our analysis shows that the reason for this is that it helps to prevent learning environmental priors. The awareness of this problem is also necessary to understand certain phenomena, for example, we hypothesize that learning environmental priors is the primary reason for the bad performance of LSTMs reported in [5], instead of the missing interaction-awareness as the authors suggest.…”

“…We further include four state-of-the-art models in our evaluation. These are the RNN-Encoder-MLP (RED) [14], the LSTM with State Refinement SR-LSTM [1] and for generative models Social GAN (S-GAN) [3] and the SoPhie GAN (SoPhie) [2]. We denote the CVM as OUR.…”

“…In particular, we analyze the influence of environmental priors, the motion history and pedestrian interactions on neural network performance. For our analysis we use the feed forward neural network (FF) and the RNN-Encoder-MLP (RED) [14] from Sec. V-B.…”

Social and Scene-Aware Trajectory Prediction in Crowded Spaces

“…For example, [20] focuses on transfer learning for pedestrian motion at intersections using Inverse Reinforcement Learning (IRL) where paths are inferred exploiting goal locations. [4] attains best performance on the challenging Stanford Drone Dataset (SDD) [17] using a recurrent-encoder and a dense layer. [22] predicts future positions in order to satisfy specific needs and to reach latent sources.…”

An RNN-Based IMM Filter Surrogate

“…Traditionally, the task of object motion prediction is done by using a Bayesian formulation in approaches such as the Kalman filter [17], or nonparametric methods, such as particle filters [4]. Driven by the success of recurrent neural networks (RNNs) in modeling temporal dependencies in a variety of sequence processing tasks, such as speech recognition [13,10] and caption generation [12,27], RNNs are increasingly utilized for object motion prediction [2,3,15,16,7]. When relying on traditional approaches, the challenge of varying dynamics over time is commonly addressed with the Interacting Multiple Model (IMM) filter [9].…”