OpenTraj: Assessing Prediction Complexity in Human Trajectories Datasets (2020-10-02T00:00:00.000000Z)

TL;DR

This paper addresses the question of evaluating how complex is a given dataset with respect to the prediction problem by defining a series of indicators around three concepts: Trajectory predictability;Trajectory regularity; Context complexity.

Abstract

Human Trajectory Prediction (HTP) has gained much momentum in the last years and many solutions have been proposed to solve it. Proper benchmarking being a key issue for comparing methods, this paper addresses the question of evaluating how complex is a given dataset with respect to the prediction problem. For assessing a dataset complexity, we define a series of indicators around three concepts: Trajectory predictability; Trajectory regularity; Context complexity. We compare the most common datasets used in HTP in the light of these indicators and discuss what this may imply on benchmarking of HTP algorithms. Our source code is released on Github.

Authors

Javad Amirian

1 papers

Bingqing Zhang

1 papers

Francisco Valente Castro

1 papers

TL;DR

Abstract

Authors

References40 items

Human Trajectory Forecasting in Crowds: A Deep Learning Perspective

Transformer Networks for Trajectory Forecasting

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

The Garden of Forking Paths: Towards Multi-Future Trajectory Prediction

Scalability in Perception for Autonomous Driving: Waymo Open Dataset

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

Argoverse: 3D Tracking and Forecasting With Rich Maps

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

nuScenes: A Multimodal Dataset for Autonomous Driving

Top-view Trajectories: A Pedestrian Dataset of Vehicle-Crowd Interaction from Controlled Experiments and Crowded Campus

Assessing spatiotemporal predictability of LBSN: a case study of three Foursquare datasets

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

Social GAN: Socially Acceptable Trajectories with Generative Adversarial Networks

Cooperative Starting Movement Detection of Cyclists Using Convolutional Neural Networks and a Boosted Stacking Ensemble

WILDTRACK: A Multi-camera HD Dataset for Dense Unscripted Pedestrian Detection

Online learning for human classification in 3D LiDAR-based tracking

Learning Social Etiquette: Human Trajectory Understanding In Crowded Scenes

Predicting Shot Making in Basketball Learnt from Adversarial Multiagent Trajectories

Social LSTM: Human Trajectory Prediction in Crowded Spaces

MOTChallenge 2015: Towards a Benchmark for Multi-Target Tracking

Universal power law governing pedestrian interactions.

The Ko-PER intersection laserscanner and video dataset

Person Tracking in Large Public Spaces Using 3-D Range Sensors

Vision meets robotics: The KITTI dataset

Collision avoidance between two walkers: role-dependent strategies.

Trajectory Extraction and Density Analysis of Intersecting Pedestrian Flows from Video Recordings

A large-scale benchmark dataset for event recognition in surveillance video

Who are you with and where are you going?

PETS2009: Dataset and challenge

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

Modelling pedestrian trajectory patterns with Gaussian processes

Depth and Appearance for Mobile Scene Analysis

Crowds by Example

New Insights into Pedestrian Flow Through Bottlenecks

Data Analysis and Graphics Using R: An Example-based Approach (Cambridge Series in Statistical and Probabilistic Mathematics)

Social force model for pedestrian dynamics.

Trajnet++: Large-scale trajectory forecasting benchmark. In: Proc. of the 2nd Workshop on Long-Term Human Motion Prediction (ICRA)

Statistical models of pedestrian behaviour in the Forum

Guiding Visual Surveillance by Tracking Human Attention

Assessing Prediction Complexity in

Field of Study

Journal Information

Name

Volume

Venue Information

Name

Type

URL

Alternate Names