Pedestrian trajectory prediction with convolutional neural networks

Zamboni, Simone; Kefato, Zekarias T.; Girdzijauskas, Šarūnas; Norén, Christoffer; Col, Laura Dal

doi:10.1016/j.patcog.2021.108252

Cited by 82 publications

(29 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…As previously stated, the suggested detection model obtains the coordinate information of the tracking object. Trajectory prediction approaches in [21][22][23][24] use the coordinate information to forecast the trajectory. the input object is identical to the object with the highest matching accuracy…”

Section: : Return the Length Of Overlapidlistmentioning

confidence: 99%

Object Recognition and Tracking in Moving Videos for Maritime Autonomous Surface Ships

Park

Ham

Kim

et al. 2022

JMSE

View full text Add to dashboard Cite

In autonomous driving technologies, a camera is necessary for establishing a path and detecting an object. Object recognition based on images from several cameras is required to detect impediments in autonomous ships. Furthermore, in order to avoid ship collisions, it is important to follow the movements of recognized ships. In this paper, we use the Singapore Maritime Dataset (SMD) and crawling image for model training. Then, we present four YOLO-based object recognition models and evaluate their performance in the maritime environment. Then, we propose a tracking algorithm to track the identified objects. Specially, in evaluation with high-motion video, the proposed tracking algorithm outperforms deep simple online and real-time tracking (DeepSORT) in terms of object tracking accuracy.

show abstract

Section: : Return the Length Of Overlapidlistmentioning

confidence: 99%

Object Recognition and Tracking in Moving Videos for Maritime Autonomous Surface Ships

Park

Ham

Kim

et al. 2022

JMSE

View full text Add to dashboard Cite

show abstract

“…Especially, a fast convolutional neural network (CNN) based model compared a 1D convolutional model with LSTM and showed improvement in temporal representation of trajectory [13]. Further, Simone et.al [14] elaborated upon the previous work by introducing novel preprocessing and data augmentation techniques to outperform other complex models. However, both the models output deterministic estimates of future state and do not quantify uncertainty.…”

Section: B Related Workmentioning

confidence: 99%

Uncertainty Estimation of Pedestrian Future Trajectory Using Bayesian Approximation

Nayak

Eskandarian

Doerzaph

2022

IEEE Open J. Intell. Transp. Syst.

View full text Add to dashboard Cite

Past research on pedestrian trajectory forecasting mainly focused on deterministic predictions which provide only point estimates of future states. These future estimates can help an autonomous vehicle plan its trajectory and avoid collision. However, under dynamic traffic scenarios, planning based on deterministic predictions is not trustworthy. Rather, estimating the uncertainty associated with the predicted states with certain level of confidence can lead to robust path planning. Hence, the authors propose to quantify this uncertainty during forecasting using stochastic approximation which deterministic approaches fail to capture. The current method is simple and applies Bayesian approximation during inference to standard neural network architectures for estimating uncertainty. The authors compared the predictions between the probabilistic neural network (NN) models with the standard deterministic models. The results indicate that the mean predicted path of probabilistic models was closer to the ground truth when compared with the deterministic prediction. Further, the effect of stochastic dropout of weights and long-term prediction on future state uncertainty has been studied. It was found that the probabilistic models produced better performance metrics like average displacement error (ADE) and final displacement error (FDE). Finally, the study has been extended to multiple datasets providing a comprehensive comparison for each model.

show abstract

“…Jejich nasazení proběhlo např. v rámci monitorování pěší dopravy v reálném čase [7], predikce trajektorie pohybu chodců [8], [9], [10], predikce rychlosti a časových odstupů na základě aktuální pozice chodců [11], řešení kolizí při interakci robotů s chodci [12] nebo zkoumání vztahů veličin fundamentálního diagramu [13]. JUNIORSTAV 2022 1.…”

Section: Popis Současného Stavuunclassified

Aproximace evakuační křivky metodou polynomické regrese

Uhlík¹,

Dobiášová²,

Kratochvílová³

2022

Juniorstav

View full text Add to dashboard Cite

Predikce průběhu evakuačního procesu v čase mimořádné události je klíčová pro následnou efektivní koordinaci evakuace a zásah složek IZS. Mikroskopické simulační nástroje jsou v současnosti nejpokročilejší metodou modelování evakuace, nicméně jejich využití v provozní fázi stavby je vzhledem k výpočetnímu času simulací a dynamickým vstupním datům neefektivní. Jedním z výstupů simulací je evakuační křivka. V článku je navrženo zjednodušení křivky aproximační metodou polynomické regrese, které demonstruje na základní úloze. Koeficienty polynomu mohou být využity při sestavování trénovací množiny modelu strojového učení pro predikci evakuačních křivek v reálném čase.

show abstract

Pedestrian trajectory prediction with convolutional neural networks

Cited by 82 publications

References 13 publications

Object Recognition and Tracking in Moving Videos for Maritime Autonomous Surface Ships

Object Recognition and Tracking in Moving Videos for Maritime Autonomous Surface Ships

Uncertainty Estimation of Pedestrian Future Trajectory Using Bayesian Approximation

Aproximace evakuační křivky metodou polynomické regrese

Contact Info

Product

Resources

About