Scipediacontent: Scipediacontent moved page Draft Content 640369520 to Xu et al 2018b

2021-02-02T06:24:56Z

Scipediacontent moved page Draft Content 640369520 to Xu et al 2018b

Scipediacontent: Created page with " == Abstract == Predicting the future location of vehicles is essential for safety-critical applications such as advanced driver assistance systems (ADAS) and autonomous driv..."

2021-02-02T06:24:52Z

Created page with " == Abstract == Predicting the future location of vehicles is essential for safety-critical applications such as advanced driver assistance systems (ADAS) and autonomous driv..."

New page

== Abstract ==

Predicting the future location of vehicles is essential for safety-critical applications such as advanced driver assistance systems (ADAS) and autonomous driving. This paper introduces a novel approach to simultaneously predict both the location and scale of target vehicles in the first-person (egocentric) view of an ego-vehicle. We present a multi-stream recurrent neural network (RNN) encoder-decoder model that separately captures both object location and scale and pixel-level observations for future vehicle localization. We show that incorporating dense optical flow improves prediction results significantly since it captures information about motion as well as appearance change. We also find that explicitly modeling future motion of the ego-vehicle improves the prediction accuracy, which could be especially beneficial in intelligent and automated vehicles that have motion planning capability. To evaluate the performance of our approach, we present a new dataset of first-person videos collected from a variety of scenarios at road intersections, which are particularly challenging moments for prediction because vehicle trajectories are diverse and dynamic.

Comment: To appear on ICRA 2019

== Original document ==

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

* [http://arxiv.org/abs/1809.07408 http://arxiv.org/abs/1809.07408]

* [http://arxiv.org/pdf/1809.07408 http://arxiv.org/pdf/1809.07408]

* [http://xplorestaging.ieee.org/ielx7/8780387/8793254/08794474.pdf?arnumber=8794474 http://xplorestaging.ieee.org/ielx7/8780387/8793254/08794474.pdf?arnumber=8794474],
: [http://dx.doi.org/10.1109/icra.2019.8794474 http://dx.doi.org/10.1109/icra.2019.8794474]

* [https://dblp.uni-trier.de/db/conf/icra/icra2019.html#YaoXCCAD19 https://dblp.uni-trier.de/db/conf/icra/icra2019.html#YaoXCCAD19],
: [https://ui.adsabs.harvard.edu/abs/2018arXiv180907408Y/abstract https://ui.adsabs.harvard.edu/abs/2018arXiv180907408Y/abstract],
: [https://academic.microsoft.com/#/detail/2968393007 https://academic.microsoft.com/#/detail/2968393007]

← Older revision	Revision as of 06:24, 2 February 2021
(No difference)

Xu et al 2018b - Revision history

Scipediacontent: Scipediacontent moved page Draft Content 640369520 to Xu et al 2018b

Scipediacontent: Created page with " == Abstract == Predicting the future location of vehicles is essential for safety-critical applications such as advanced driver assistance systems (ADAS) and autonomous driv..."