Scipediacontent: Scipediacontent moved page Draft Content 966838411 to Tome et al 2016a

2021-01-28T16:59:45Z

Scipediacontent moved page Draft Content 966838411 to Tome et al 2016a

Scipediacontent: Created page with " == Abstract == urate pedestrian detection has a primary role in automotive safety: for example, by issuing warnings to the driver or acting actively on car's brakes, it help..."

2021-01-28T16:59:42Z

Created page with " == Abstract == urate pedestrian detection has a primary role in automotive safety: for example, by issuing warnings to the driver or acting actively on car's brakes, it help..."

New page

== Abstract ==

urate pedestrian detection has a primary role in automotive safety: for example, by issuing warnings to the driver or acting actively on car's brakes, it helps decreasing the probability of injuries and human fatalities. In order to achieve very high accuracy, recent pedestrian detectors have been based on Convolutional Neural Networks (CNN). Unfortunately, such approaches require vast amounts of computational power and memory, preventing efficient implementations on embedded systems. This work proposes a CNN-based detector, adapting a general-purpose convolutional network to the task at hand. By thoroughly analyzing and optimizing each step of the detection pipeline, we develop an architecture that outperforms methods based on traditional image features and achieves an accuracy close to the state-of-the-art while having low computational complexity. Furthermore, the model is compressed in order to fit the tight constrains of low power devices with a limited amount of embedded memory available. This paper makes two main contributions: (1) it proves that a region based deep neural network can be finely tuned to achieve adequate accuracy for pedestrian detection (2) it achieves a very low memory usage without reducing detection accuracy on the Caltech Pedestrian dataset.

Comment: IEEE 2016 ICCE-Berlin

== Original document ==

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

* [http://arxiv.org/abs/1609.02500 http://arxiv.org/abs/1609.02500]

* [http://arxiv.org/pdf/1609.02500 http://arxiv.org/pdf/1609.02500]

* [http://xplorestaging.ieee.org/ielx7/7590260/7684690/07684706.pdf?arnumber=7684706 http://xplorestaging.ieee.org/ielx7/7590260/7684690/07684706.pdf?arnumber=7684706],
: [http://dx.doi.org/10.1109/icce-berlin.2016.7684706 http://dx.doi.org/10.1109/icce-berlin.2016.7684706]

* [https://dblp.uni-trier.de/db/journals/corr/corr1609.html#TomeBPBPT16 https://dblp.uni-trier.de/db/journals/corr/corr1609.html#TomeBPBPT16],
: [https://arxiv.org/abs/1609.02500 https://arxiv.org/abs/1609.02500],
: [https://arxiv.org/pdf/1609.02500.pdf https://arxiv.org/pdf/1609.02500.pdf],
: [https://ui.adsabs.harvard.edu/abs/2016arXiv160902500T/abstract https://ui.adsabs.harvard.edu/abs/2016arXiv160902500T/abstract],
: [https://ieeexplore.ieee.org/document/7684706 https://ieeexplore.ieee.org/document/7684706],
: [https://uk.arxiv.org/pdf/1609.02500 https://uk.arxiv.org/pdf/1609.02500],
: [http://export.arxiv.org/pdf/1609.02500 http://export.arxiv.org/pdf/1609.02500],
: [https://uk.arxiv.org/abs/1609.02500 https://uk.arxiv.org/abs/1609.02500],
: [https://export.arxiv.org/abs/1609.02500 https://export.arxiv.org/abs/1609.02500],
: [https://academic.microsoft.com/#/detail/2518858372 https://academic.microsoft.com/#/detail/2518858372]

← Older revision	Revision as of 16:59, 28 January 2021
(No difference)

Tome et al 2016a - Revision history

Scipediacontent: Scipediacontent moved page Draft Content 966838411 to Tome et al 2016a

Scipediacontent: Created page with " == Abstract == urate pedestrian detection has a primary role in automotive safety: for example, by issuing warnings to the driver or acting actively on car's brakes, it help..."