Scipediacontent: Scipediacontent moved page Draft Content 819806451 to Ross et al 2019a

2021-01-28T23:54:35Z

Scipediacontent moved page Draft Content 819806451 to Ross et al 2019a

Scipediacontent: Created page with " == Abstract == Reliable anticipation of pedestrian trajectory is imperative for the operation of autonomous vehicles and can significantly enhance the functionality of advan..."

2021-01-28T23:54:31Z

Created page with " == Abstract == Reliable anticipation of pedestrian trajectory is imperative for the operation of autonomous vehicles and can significantly enhance the functionality of advan..."

New page

== Abstract ==

Reliable anticipation of pedestrian trajectory is imperative for the operation of autonomous vehicles and can significantly enhance the functionality of advanced driver assistance systems. While significant progress has been made in the field of pedestrian detection, forecasting pedestrian trajectories remains a challenging problem due to the unpredictable nature of pedestrians and the huge space of potentially useful features. In this work, we present a deep learning approach for pedestrian trajectory forecasting using a single vehicle-mounted camera. Deep learning models that have revolutionized other areas in computer vision have seen limited application to trajectory forecasting, in part due to the lack of richly annotated training data. We address the lack of training data by introducing a scalable machine annotation scheme that enables our model to be trained using a large dataset without human annotation. In addition, we propose Dynamic Trajectory Predictor (DTP), a model for forecasting pedestrian trajectory up to one second into the future. DTP is trained using both human and machine-annotated data, and anticipates dynamic motion that is not captured by linear models. Experimental evaluation confirms the benefits of the proposed model.

Comment: 6 pages, 5 figures. To appear in the proceedings of the 2019 IEEE Intelligent Vehicles Symposium (IV)

== Original document ==

The different versions of the original document can be found in:

* [http://arxiv.org/abs/1905.03681 http://arxiv.org/abs/1905.03681]

* [http://arxiv.org/pdf/1905.03681 http://arxiv.org/pdf/1905.03681]

* [http://dx.doi.org/10.1109/ivs.2019.8814207 http://dx.doi.org/10.1109/ivs.2019.8814207]

* [http://wrap.warwick.ac.uk/117169/1/WRAP-forecasting-pedestrian-trajectory-machine-annotated-training-data-Styles-2019.pdf http://wrap.warwick.ac.uk/117169/1/WRAP-forecasting-pedestrian-trajectory-machine-annotated-training-data-Styles-2019.pdf]

* [http://xplorestaging.ieee.org/ielx7/8792328/8813768/08814207.pdf?arnumber=8814207 http://xplorestaging.ieee.org/ielx7/8792328/8813768/08814207.pdf?arnumber=8814207],
: [http://dx.doi.org/10.1109/ivs.2019.8814207 http://dx.doi.org/10.1109/ivs.2019.8814207]

* [https://dblp.uni-trier.de/db/journals/corr/corr1905.html#abs-1905-03681 https://dblp.uni-trier.de/db/journals/corr/corr1905.html#abs-1905-03681],
: [https://ieeexplore.ieee.org/document/8814207 https://ieeexplore.ieee.org/document/8814207],
: [http://wrap.warwick.ac.uk/117169 http://wrap.warwick.ac.uk/117169],
: [https://academic.microsoft.com/#/detail/2970733097 https://academic.microsoft.com/#/detail/2970733097]

← Older revision	Revision as of 23:54, 28 January 2021
(No difference)

Ross et al 2019a - Revision history

Scipediacontent: Scipediacontent moved page Draft Content 819806451 to Ross et al 2019a

Scipediacontent: Created page with " == Abstract == Reliable anticipation of pedestrian trajectory is imperative for the operation of autonomous vehicles and can significantly enhance the functionality of advan..."