Early Start Intention Detection of Cyclists Using Motion History Images and a Deep Residual Network

Zernetsch, Stefan; Kress, Viktor; Sick, Bernhard; Doll, Konrad

doi:10.1109/ivs.2018.8500428

Cited by 15 publications

(7 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Moreover, we demonstrated that training distinct classifiers for specific locations can further improve detection results, i.e., reaching an F 1 -score of 94% with a mean detection time of 0.34 s for the device worn in the trouser pocket. Our proposed approach is still not fully competitive with accurate video-based methods (e.g., [5]) as available in infrastructure and vehicles, yet. Nevertheless, these methods fail in case of bad visibility or occlusion.…”

Section: Discussionmentioning

confidence: 86%

“…The approaches for pedestrian intention detection are based on recursive Bayesian filter (e.g., Kalman filter) and machine learning techniques, e.g., Gaussian process dynamical models, dynamic Bayesian networks, and probabilistic hierarchical trajectory matching. In [5], Zernetsch et al use machine learning techniques, i.e., support-vector machines and convolutional neural networks, to image sequences of an infrastructure-based camera system for early starting intention detection of cyclists. Their approach safely detects starting movements on average 0.14 s after the cyclist starts moving.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Starting Movement Detection of Cyclists Using Smart Devices

Bieshaar

Depping

Schneegans

et al. 2018

2018 IEEE 5th International Conference on Data Science and Advanced Analytics (DSAA)

Self Cite

View full text Add to dashboard Cite

In near future, vulnerable road users (VRUs) such as cyclists and pedestrians will be equipped with smart devices and wearables which are capable to communicate with intelligent vehicles and other traffic participants. Road users are then able to cooperate on different levels, such as in cooperative intention detection for advanced VRU protection. Smart devices can be used to detect intentions, e.g., an occluded cyclist intending to cross the road, to warn vehicles of VRUs, and prevent potential collisions. This article presents a human activity recognition approach to detect the starting movement of cyclists wearing smart devices. We propose a novel two-stage feature selection procedure using a score specialized for robust starting detection reducing the false positive detections and leading to understandable and interpretable features. The detection is modelled as a classification problem and realized by means of a machine learning classifier. We introduce an auxiliary class, that models starting movements and allows to integrate early movement indicators, i.e., body part movements indicating future behaviour. In this way we improve the robustness and reduce the detection time of the classifier. Our empirical studies with real-world data originating from experiments which involve 49 test subjects and consists of 84 starting motions show that we are able to detect the starting movements early. Our approach reaches an F1 F1 F1-score of 67% within 0.33 s after the first movement of the bicycle wheel. Investigations concerning the device wearing location show that for devices worn in the trouser pocket the detector has less false detections and detects starting movements faster on average. We found that we can further improve the results when we train distinct classifiers for different wearing locations. In this case we reach an F1 F1 F1-score of 94% with a mean detection time of 0.34 s for the device worn in the trouser pocket.

show abstract

Section: Discussionmentioning

confidence: 86%

Section: Related Workmentioning

confidence: 99%

Starting Movement Detection of Cyclists Using Smart Devices

Bieshaar

Depping

Schneegans

et al. 2018

2018 IEEE 5th International Conference on Data Science and Advanced Analytics (DSAA)

Self Cite

View full text Add to dashboard Cite

show abstract

“…In both cases, the MHI based approach outperformed an approach based on a Kalman filter with interacting multiple models. In [8], the MHI based approach was adapted for cyclist starting movement detection and extended by a residual convolutional neural network (ResNet), where the ResNet outperformed the SVM approach. In this article, the approach from [8] is adapted to a wide angle stereo camera system and extended to cover all possible movement types of cyclists.…”

Section: B Related Workmentioning

confidence: 99%

Cyclist Intention Detection: A Probabilistic Approach

Zernetsch¹,

Reichert²,

Kress³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

This article presents a holistic approach for probabilistic cyclist intention detection. A basic movement detection based on motion history images (MHI) and a residual convolutional neural network (ResNet) are used to estimate probabilities for the current cyclist motion state. These probabilities are used as weights in a probabilistic ensemble trajectory forecast. The ensemble consists of specialized models, which produce individual forecasts in the form of Gaussian distributions under the assumption of a certain motion state of the cyclist (e.g. cyclist is starting or turning left). By weighting the specialized models, we create forecasts in the from of Gaussian mixtures that define regions within which the cyclists will reside with a certain probability. To evaluate our method, we rate the reliability, sharpness, and positional accuracy of our forecasted distributions. We compare our method to a single model approach which produces forecasts in the form of Gaussian distributions and show that our method is able to produce more reliable and sharper outputs while retaining comparable positional accuracy. Both methods are evaluated using a dataset created at a public traffic intersection. Our code and the dataset are made publicly available.

show abstract

“…A closely related field that also uses context cues is intent prediction for VRUs. Here, context cues can also be used to predict the intent of a VRU, such as the pose [13], or do the intention prediction directly on the image data [14].…”

Section: Previous Workmentioning

confidence: 99%

Context-based cyclist path prediction using Recurrent Neural Networks

Pool

Kooij

Gavrila

2019

2019 IEEE Intelligent Vehicles Symposium (IV)

View full text Add to dashboard Cite

This paper proposes a Recurrent Neural Network (RNN) for cyclist path prediction to learn the effect of contextual cues on the behavior directly in an end-to-end approach, removing the need for any annotations. The proposed RNN incorporates three distinct contextual cues: one related to actions of the cyclist, one related to the location of the cyclist on the road, and one related to the interaction between the cyclist and the egovehicle. The RNN predicts a Gaussian distribution over the future position of the cyclist one second into the future with a higher accuracy, compared to a current state-of-the-art model that is based on dynamic mode annotations, where our model attains an average prediction error of 33 cm one second into the future.

show abstract

Early Start Intention Detection of Cyclists Using Motion History Images and a Deep Residual Network

Cited by 15 publications

References 18 publications

Starting Movement Detection of Cyclists Using Smart Devices

Starting Movement Detection of Cyclists Using Smart Devices

Cyclist Intention Detection: A Probabilistic Approach

Context-based cyclist path prediction using Recurrent Neural Networks

Contact Info

Product

Resources

About