The Atlas Benchmark: an Automated Evaluation Framework for Human Motion Prediction

被引:3
作者
Rudenko, Andrey [1 ]
Palmieri, Luigi [1 ]
Huang, Wanting [2 ]
Lilienthal, Achim J. [3 ]
Arras, Kai O. [1 ]
机构
[1] Robert Bosch GmbH, Corp Res, Stuttgart, Germany
[2] Tech Univ Munich, Munich, Germany
[3] Orebro Univ Sweden, Mobile Robot & Olfact Lab, Mobile, AL, Sweden
来源
2022 31ST IEEE INTERNATIONAL CONFERENCE ON ROBOT AND HUMAN INTERACTIVE COMMUNICATION (IEEE RO-MAN 2022) | 2022年
关键词
MODEL;
D O I
10.1109/RO-MAN53752.2022.9900656
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
Human motion trajectory prediction, an essential task for autonomous systems in many domains, has been on the rise in recent years. With a multitude of new methods proposed by different communities, the lack of standardized benchmarks and objective comparisons is increasingly becoming a major limitation to assess progress and guide further research. Existing benchmarks are limited in their scope and flexibility to conduct relevant experiments and to account for contextual cues of agents and environments. In this paper we present Atlas, a benchmark to systematically evaluate human motion trajectory prediction algorithms in a unified framework. Atlas offers data preprocessing functions, hyperparameter optimization, comes with popular datasets and has the flexibility to setup and conduct underexplored yet relevant experiments to analyze a method's accuracy and robustness. In an example application of Atlas, we compare five popular model- and learning-based predictors and find that, when properly applied, early physics-based approaches are still remarkably competitive. Such results confirm the necessity of benchmarks like Atlas.
引用
收藏
页码:636 / 643
页数:8
相关论文
共 39 条
[1]   Social LSTM: Human Trajectory Prediction in Crowded Spaces [J].
Alahi, Alexandre ;
Goel, Kratarth ;
Ramanathan, Vignesh ;
Robicquet, Alexandre ;
Li Fei-Fei ;
Savarese, Silvio .
2016 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2016, :961-971
[2]   Social Ways: Learning Multi-Modal Distributions of Pedestrian Trajectories with GANs [J].
Amirian, Javad ;
Hayet, Jean-Bernard ;
Pettre, Julien .
2019 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION WORKSHOPS (CVPRW 2019), 2019, :2964-2972
[3]  
Bartoli F, 2018, INT C PATT RECOG, P1941, DOI 10.1109/ICPR.2018.8545447
[4]  
Becker S, 2018, PROC EUROP C COMP VI
[5]   Person Tracking in Large Public Spaces Using 3-D Range Sensors [J].
Brscic, Drazen ;
Kanda, Takayuki ;
Ikeda, Tetsushi ;
Miyashita, Takahiro .
IEEE TRANSACTIONS ON HUMAN-MACHINE SYSTEMS, 2013, 43 (06) :522-534
[6]   Walking Ahead: The Headed Social Force Model [J].
Farina, Francesco ;
Fontanelli, Daniele ;
Garulli, Andrea ;
Giannitrapani, Antonio ;
Prattichizzo, Domenico .
PLOS ONE, 2017, 12 (01)
[7]   Soft plus Hardwired attention: An LSTM framework for human trajectory prediction and abnormal event detection [J].
Fernando, Tharindu ;
Denman, Simon ;
Sridharan, Sridha ;
Fookes, Clinton .
NEURAL NETWORKS, 2018, 108 :466-478
[8]   Social GAN: Socially Acceptable Trajectories with Generative Adversarial Networks [J].
Gupta, Agrim ;
Johnson, Justin ;
Li Fei-Fei ;
Savarese, Silvio ;
Alahi, Alexandre .
2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2018, :2255-2264
[9]   Bench-MR: A Motion Planning Benchmark for Wheeled Mobile Robots [J].
Heiden, Eric ;
Palmieri, Luigi ;
Bruns, Leonard ;
Arras, Kai O. ;
Sukhatme, Gaurav S. ;
Koenig, Sven .
IEEE ROBOTICS AND AUTOMATION LETTERS, 2021, 6 (03) :4536-4543
[10]   SOCIAL FORCE MODEL FOR PEDESTRIAN DYNAMICS [J].
HELBING, D ;
MOLNAR, P .
PHYSICAL REVIEW E, 1995, 51 (05) :4282-4286