Exploring Dynamic Context for Multi-path Trajectory Prediction (2020-10-30T00:00:00.000000Z)

TL;DR

In this paper, a novel framework, named Dynamic Context Encoder Network (DCENet), the spatial context between agents is explored by using self-attention architectures, and a set of future trajectories for each agent is predicted conditioned on the learned spatial-temporal context by sampling from the latent space.

Abstract

To accurately predict future positions of different agents in traffic scenarios is crucial for safely deploying intelligent autonomous systems in the real-world environment. However, it remains a challenge due to the behavior of a target agent being affected by other agents dynamically and there being more than one socially possible paths the agent could take. In this paper, we propose a novel framework, named Dynamic Context Encoder Network (DCENet). In our framework, first, the spatial context between agents is explored by using self-attention architectures. Then, the two-stream encoders are trained to learn temporal context between steps by taking the respective observed trajectories and the extracted dynamic spatial context as input. The spatial-temporal context is encoded into a latent space using a Conditional Variational Auto-Encoder (CVAE) module. Finally, a set of future trajectories for each agent is predicted conditioned on the learned spatial-temporal context by sampling from the latent space, repeatedly. DCENet is evaluated on one of the most popular challenging benchmarks for trajectory forecasting Trajnet and reports a new state-of-the-art performance. It also demonstrates superior performance evaluated on the benchmark inD for mixed traffic at intersections. A series of ablation studies is conducted to validate the effectiveness of each proposed module. Our code is available at https://github.com/wtliao/DCENet.

Authors

Monika Sester

2 papers

M. Yang

8 papers

B. Rosenhahn

9 papers

TL;DR

Abstract

Authors

References53 items

Generative Adversarial Networks

Human Trajectory Forecasting in Crowds: A Deep Learning Perspective

AMENet: Attentive Maps Encoder Network for Trajectory Prediction

Intent-Aware Pedestrian Prediction for Adaptive Crowd Navigation

Image Captioning through Image Transformer

M2P3: multimodal multi-pedestrian path prediction by self-driving cars with egocentric vision

Transformer Networks for Trajectory Forecasting

MCENET: Multi-Context Encoder Network for Homogeneous Agent Trajectory Prediction in Mixed Traffic

Trajectron++: Dynamically-Feasible Trajectory Forecasting with Heterogeneous Data

Forecasting Trajectory and Behavior of Road-Agents Using Spectral Clustering in Graph-LSTMs

The inD Dataset: A Drone Dataset of Naturalistic Road User Trajectories at German Intersections

Multiple Futures Prediction

Diverse Trajectory Forecasting with Determinantal Point Processes

Overcoming Limitations of Mixture Density Networks: A Sampling and Fitting Framework for Multimodal Future Prediction

Social Ways: Learning Multi-Modal Distributions of Pedestrian Trajectories With GANs

SR-LSTM: State Refinement for LSTM Towards Pedestrian Trajectory Prediction

Multi-Step Prediction of Occupancy Grid Maps With Recurrent Neural Networks

TraPHic: Trajectory Prediction in Dense and Heterogeneous Traffic Using Weighted Interactions

IntentNet: Learning to Predict Intention from Raw Sensor Data

Multimodal Trajectory Predictions for Autonomous Driving using Deep Convolutional Networks

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

Move, Attend and Predict: An attention-based neural model for people's movement prediction

SoPhie: An Attentive GAN for Predicting Paths Compliant to Social and Physical Constraints

An Evaluation of Trajectory Prediction Approaches and Notes on the TrajNet Benchmark

MX-LSTM: Mixing Tracklets and Vislets to Jointly Forecast Trajectories and Head Poses

Social GAN: Socially Acceptable Trajectories with Generative Adversarial Networks

Social Attention: Modeling Attention in Human Crowds

Bottom-Up and Top-Down Attention for Image Captioning and Visual Question Answering

Attention is All you Need

DESIRE: Distant Future Prediction in Dynamic Scenes with Interacting Agents

Learning Social Etiquette: Human Trajectory Understanding In Crowded Scenes

Social LSTM: Human Trajectory Prediction in Crowded Spaces

Learning Structured Output Representation using Deep Conditional Generative Models

Neural Machine Translation by Jointly Learning to Align and Translate

Semi-supervised Learning with Deep Generative Models

Two-Stream Convolutional Networks for Action Recognition in Videos

Auto-Encoding Variational Bayes

Generating Sequences With Recurrent Neural Networks

Multiple Choice Learning: Learning to Produce Multiple Structured Outputs

PETS2009: Dataset and challenge

You'll never walk alone: Modeling social behavior for multi-target tracking

Modelling pedestrian trajectory patterns with Gaussian processes

Crowds by Example

The negotiation of stationary and moving obstructions during walking: anticipatory locomotor adaptations and preservation of personal space.

Spatial and Probabilistic Modelling of Pedestrian Behaviour

Social force model for pedestrian dynamics.

Forecasting structural time series models and the kalman filter: Andrew Harvey, 1989, (Cambridge University Press), 554 pp., ISBN 0-521-32196-4

Forecasting, Structural Time Series Models and the Kalman Filter.

Sur les fonctions convexes et les inégalités entre les valeurs moyennes

BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

Deep Learning

Research Showcase @ CMU

Modelling Smooth Paths Using Gaussian Processes

Field of Study

Journal Information

Name

Page

Venue Information

Name

Type

URL

Alternate Names