Transformer Networks for Trajectory Forecasting

by   Francesco Giuliari, et al.
University of Verona
Sapienza University of Rome

Most recent successes on forecasting the people motion are based on LSTM models and all most recent progress has been achieved by modelling the social interaction among people and the people interaction with the scene. We question the use of the LSTM models and propose the novel use of Transformer Networks for trajectory forecasting. This is a fundamental switch from the sequential step-by-step processing of LSTMs to the only-attention-based memory mechanisms of Transformers. In particular, we consider both the original Transformer Network (TF) and the larger Bidirectional Transformer (BERT), state-of-the-art on all natural language processing tasks. Our proposed Transformers predict the trajectories of the individual people in the scene. These are " model because each person is modelled separately without any complex human-human nor scene interaction terms. In particular, the TF model without bells and whistles yields the best score on the largest and most challenging trajectory forecasting benchmark of TrajNet. Additionally, its extension which predicts multiple plausible future trajectories performs on par with more engineered techniques on the 5 datasets of ETH+UCY. Finally, we show that Transformers may deal with missing observations, as it may be the case with real sensor data.


Under the Hood of Transformer Networks for Trajectory Forecasting

Transformer Networks have established themselves as the de-facto state-o...

Multimodal Vision Transformers with Forced Attention for Behavior Analysis

Human behavior understanding requires looking at minute details in the l...

Forecasting People Trajectories and Head Poses by Jointly Reasoning on Tracklets and Vislets

In this work, we explore the correlation between people trajectories and...

Uncertainty-aware State Space Transformer for Egocentric 3D Hand Trajectory Forecasting

Hand trajectory forecasting from egocentric views is crucial for enablin...

Forecasting Human Trajectory from Scene History

Predicting the future trajectory of a person remains a challenging probl...

Latent Variable Nested Set Transformers AutoBots

Humans have the innate ability to attend to the most relevant actors in ...

Goal-driven Self-Attentive Recurrent Networks for Trajectory Prediction

Human trajectory forecasting is a key component of autonomous vehicles, ...

Please sign up or login with your details

Forgot password? Click here to reset