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

Schöller, Christoph; Aravantinos, Vincent; Lay, Florian; Knoll, Alois

doi:10.48550/arxiv.1903.07933

Cited by 4 publications

(9 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Generative adversarial networks [27] are used in [2], [3] to learn to predict multiple trajectories. [10] show that those predictions do not capture the multi-modal nature of the trajectories. For social awareness, social interactions are represented through a grid-based map of their local neighbourhood and fed into a Recurrent Neural Network (RNN) [1], [2], [28], [29].…”

Section: Related Workmentioning

confidence: 91%

“…Three baseline models that do not use social information are studied along with the four social models. First baseline is the Constant velocity(CV) presented in [10]. The second baseline from [9] is an MLP fed with features extracted from the motion history of a pedestrian using a LSTM to predict the pedestrian's future path, referred to as LSTM-MLP.…”

Section: Methodology a Baselinesmentioning

confidence: 99%

“…The same architecture is used in [9], but with a long short-term memory (LSTM) encoder instead. Surprisingly, the authors of [10] show that a simple constant-velocity approach could outperform more complex approaches such as those presented in [2], [3].…”

Section: Related Workmentioning

confidence: 99%

“…Using the groundtruth trajectory to select the best predicted trajectory is not a correct practice and we therefore prefer to avoid it. Furthermore, [10] pointed out that it is similar as generating a set of constant velocity trajectories with a random angle and selecting the best one. The models are trained using the Mean Squared Error between the predicted and groundtruth trajectories as loss function through backpropagation as in social-lstm [1], vain [5] or cidnn [33].…”

Section: Related Workmentioning

confidence: 99%

“…models not taking social information as input, could outperform social models on both metrics. Moreover, a simple constant velocity model is shown in [10] to perform better than most neural networks based models, including social models. This invalidates the previous interpretations of ablative studies.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Soft Attention: Does it Actually Help to Learn Social Interactions in Pedestrian Trajectory Prediction?

Boucaud,

Aloise,

Saunier

2021

Preprint

View full text Add to dashboard Cite

We consider the problem of predicting the future path of a pedestrian using its motion history and the motion history of the surrounding pedestrians, called social information. Since the seminal paper on Social-LSTM, deep-learning has become the main tool used to model the impact of social interactions on a pedestrian's motion. The demonstration that these models can learn social interactions relies on an ablative study of these models. The models are compared with and without their social interactions module on two standard metrics, the Average Displacement Error and Final Displacement Error. Yet, these complex models were recently outperformed by a simple constant-velocity approach. This questions if they actually allow to model social interactions as well as the validity of the proof. In this paper, we focus on the deep-learning models with a soft-attention mechanism for social interaction modeling and study whether they use social information at prediction time. We conduct two experiments across four state-of-the-art approaches on the ETH and UCY datasets, which were also used in previous work. First, the models are trained by replacing the social information with random noise and compared to model trained with actual social information. Second, we use a gating mechanism along with a L0 penalty, allowing models to shut down their inner components. The models consistently learn to prune their soft-attention mechanism. For both experiments, neither the course of the convergence nor the prediction performance were altered. This demonstrates that the soft-attention mechanism and therefore the social information are ignored by the models.

show abstract

Section: Related Workmentioning

confidence: 91%

Section: Methodology a Baselinesmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Soft Attention: Does it Actually Help to Learn Social Interactions in Pedestrian Trajectory Prediction?

Boucaud,

Aloise,

Saunier

2021

Preprint

View full text Add to dashboard Cite

show abstract

On Adversarial Robustness of Trajectory Prediction for Autonomous Vehicles

Zhang

Sun

et al. 2022

2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

View full text Add to dashboard Cite

Autonomous vehicles rely on accurate trajectory prediction to inform decision-making processes related to navigation and collision avoidance. However, current trajectory prediction models show signs of overfitting, which may lead to unsafe or suboptimal behavior. To address these challenges, this paper presents a comprehensive framework that categorizes and assesses the definitions and strategies used in the literature on evaluating and improving the robustness of trajectory prediction models. This involves a detailed exploration of various approaches, including data slicing methods, perturbation techniques, model architecture changes, and post-training adjustments. In the literature, we see many promising methods for increasing robustness, which are necessary for safe and reliable autonomous driving.

show abstract

End-to-end, real time and robust behavioral prediction module with ROS for autonomous vehicles

Kayın,

Erdaş

2024

Communications Faculty of Sciences University of Ankara Series A2-A3 Physical Sciences and Engineering

View full text Add to dashboard Cite

In the world where urbanization and population density are increasing, transportation methods are also diversifying and the use of unmanned vehicles is becoming widespread. In order for unmanned vehicles to perform their tasks autonomously, they need to be able to perceive their own position, the environment and predict the possible movements/routes of environmental factors, similar to living things. In autonomous vehicles, it is extremely important for the safety of the vehicle and the surrounding factors to be able to predict the future position of the objects around it with high performance so that the vehicle can plan correctly. Due to the stated reasons, the behavioral prediction module is a very important component for autonomous vehicles, especially in moving environments. In this study, fast and successful robotic behavioral prediction module has been developed to enable the autonomous vehicle to plan more safely and successfully.

show abstract

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

Cited by 4 publications

References 0 publications

Soft Attention: Does it Actually Help to Learn Social Interactions in Pedestrian Trajectory Prediction?

Soft Attention: Does it Actually Help to Learn Social Interactions in Pedestrian Trajectory Prediction?

On Adversarial Robustness of Trajectory Prediction for Autonomous Vehicles

End-to-end, real time and robust behavioral prediction module with ROS for autonomous vehicles

Contact Info

Product

Resources

About