Scipediacontent: Scipediacontent moved page Draft Content 462074658 to Ruzicka Franchetti 2018a

2021-02-03T17:51:40Z

Scipediacontent moved page Draft Content 462074658 to Ruzicka Franchetti 2018a

Scipediacontent: Created page with " == Abstract == Machine learning has celebrated a lot of achievements on computer vision tasks such as object detection, but the traditionally used models work with relativel..."

2021-02-03T17:51:37Z

Created page with " == Abstract == Machine learning has celebrated a lot of achievements on computer vision tasks such as object detection, but the traditionally used models work with relativel..."

New page

== Abstract ==

Machine learning has celebrated a lot of achievements on computer vision tasks such as object detection, but the traditionally used models work with relatively low resolution images. The resolution of recording devices is gradually increasing and there is a rising need for new methods of processing high resolution data. We propose an attention pipeline method which uses two staged evaluation of each image or video frame under rough and refined resolution to limit the total number of necessary evaluations. For both stages, we make use of the fast object detection model YOLO v2. We have implemented our model in code, which distributes the work across GPUs. We maintain high accuracy while reaching the average performance of 3-6 fps on 4K video and 2 fps on 8K video.

Comment: 6 pages, 12 figures, Best Paper Finalist at IEEE High Performance Extreme Computing Conference (HPEC) 2018; copyright 2018 IEEE; (DOI will be filled when known)

== Original document ==

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

* [http://arxiv.org/abs/1810.10551 http://arxiv.org/abs/1810.10551]

* [http://arxiv.org/pdf/1810.10551 http://arxiv.org/pdf/1810.10551]

* [http://xplorestaging.ieee.org/ielx7/8534508/8547513/08547574.pdf?arnumber=8547574 http://xplorestaging.ieee.org/ielx7/8534508/8547513/08547574.pdf?arnumber=8547574],
: [http://dx.doi.org/10.1109/hpec.2018.8547574 http://dx.doi.org/10.1109/hpec.2018.8547574]

* [https://dblp.uni-trier.de/db/journals/corr/corr1810.html#abs-1810-10551 https://dblp.uni-trier.de/db/journals/corr/corr1810.html#abs-1810-10551],
: [https://arxiv.org/pdf/1810.10551.pdf https://arxiv.org/pdf/1810.10551.pdf],
: [https://arxiv.org/abs/1810.10551 https://arxiv.org/abs/1810.10551],
: [https://www.arxiv-vanity.com/papers/1810.10551 https://www.arxiv-vanity.com/papers/1810.10551],
: [http://export.arxiv.org/pdf/1810.10551 http://export.arxiv.org/pdf/1810.10551],
: [https://uk.arxiv.org/abs/1810.10551?context=cs https://uk.arxiv.org/abs/1810.10551?context=cs],
: [https://ru.arxiv.org/abs/1810.10551 https://ru.arxiv.org/abs/1810.10551],
: [https://ru.arxiv.org/pdf/1810.10551 https://ru.arxiv.org/pdf/1810.10551],
: [https://export.arxiv.org/abs/1810.10551 https://export.arxiv.org/abs/1810.10551],
: [https://academic.microsoft.com/#/detail/2898291770 https://academic.microsoft.com/#/detail/2898291770]

← Older revision	Revision as of 17:51, 3 February 2021
(No difference)

Ruzicka Franchetti 2018a - Revision history

Scipediacontent: Scipediacontent moved page Draft Content 462074658 to Ruzicka Franchetti 2018a

Scipediacontent: Created page with " == Abstract == Machine learning has celebrated a lot of achievements on computer vision tasks such as object detection, but the traditionally used models work with relativel..."