Scipediacontent: Scipediacontent moved page Draft Content 837906997 to Xu et al 2019b

2021-02-02T00:27:40Z

Scipediacontent moved page Draft Content 837906997 to Xu et al 2019b

Scipediacontent: Created page with " == Abstract == Recognizing abnormal events such as traffic violations and accidents in natural driving scenes is essential for successful autonomous driving and advanced dri..."

2021-02-02T00:27:33Z

Created page with " == Abstract == Recognizing abnormal events such as traffic violations and accidents in natural driving scenes is essential for successful autonomous driving and advanced dri..."

New page

== Abstract ==

Recognizing abnormal events such as traffic violations and accidents in natural driving scenes is essential for successful autonomous driving and advanced driver assistance systems. However, most work on video anomaly detection suffers from two crucial drawbacks. First, they assume cameras are fixed and videos have static backgrounds, which is reasonable for surveillance applications but not for vehicle-mounted cameras. Second, they pose the problem as one-class classification, relying on arduously hand-labeled training datasets that limit recognition to anomaly categories that have been explicitly trained. This paper proposes an unsupervised approach for traffic accident detection in first-person (dashboard-mounted camera) videos. Our major novelty is to detect anomalies by predicting the future locations of traffic participants and then monitoring the prediction accuracy and consistency metrics with three different strategies. We evaluate our approach using a new dataset of diverse traffic accidents, AnAn Accident Detection (A3D), as well as another publicly-available dataset. Experimental results show that our approach outperforms the state-of-the-art.

Comment: Accepted to IROS 2019

== Original document ==

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

* [http://arxiv.org/abs/1903.00618 http://arxiv.org/abs/1903.00618]

* [http://arxiv.org/pdf/1903.00618 http://arxiv.org/pdf/1903.00618]

* [http://xplorestaging.ieee.org/ielx7/8957008/8967518/08967556.pdf?arnumber=8967556 http://xplorestaging.ieee.org/ielx7/8957008/8967518/08967556.pdf?arnumber=8967556],
: [http://dx.doi.org/10.1109/iros40897.2019.8967556 http://dx.doi.org/10.1109/iros40897.2019.8967556]

* [https://dblp.uni-trier.de/db/conf/iros/iros2019.html#YaoXWCA19 https://dblp.uni-trier.de/db/conf/iros/iros2019.html#YaoXWCA19],
: [https://arxiv.org/abs/1903.00618 https://arxiv.org/abs/1903.00618],
: [https://arxiv.org/pdf/1903.00618.pdf https://arxiv.org/pdf/1903.00618.pdf],
: [https://doi.org/10.1109/IROS40897.2019.8967556 https://doi.org/10.1109/IROS40897.2019.8967556],
: [https://academic.microsoft.com/#/detail/3003987610 https://academic.microsoft.com/#/detail/3003987610]

← Older revision	Revision as of 00:27, 2 February 2021
(No difference)

Xu et al 2019b - Revision history

Scipediacontent: Scipediacontent moved page Draft Content 837906997 to Xu et al 2019b

Scipediacontent: Created page with " == Abstract == Recognizing abnormal events such as traffic violations and accidents in natural driving scenes is essential for successful autonomous driving and advanced dri..."