Vehicle Counting using Video Metadata
(2018) In LU-CS-EX 2018-13 EDAM05 20181Department of Computer Science
- Abstract
- The current field of object detection and image recognition is huge but not
without complications. Processing large amounts of high resolution videos
needs powerful hardware and also risks breaching the privacy of those who
are recorded. In times of increasing demand for decentralized solutions and
stricter privacy protection regulations being put in place a new approach is
needed.
We present an alternative to traditional object detection in video where we
analyze changes to its metadata over time rather than the content of the video
frames. This approach has several benefits over traditional object detection: it
is incredibly fast, lightweight and protects the privacy of its subjects.
We have trained and evaluated several... (More) - The current field of object detection and image recognition is huge but not
without complications. Processing large amounts of high resolution videos
needs powerful hardware and also risks breaching the privacy of those who
are recorded. In times of increasing demand for decentralized solutions and
stricter privacy protection regulations being put in place a new approach is
needed.
We present an alternative to traditional object detection in video where we
analyze changes to its metadata over time rather than the content of the video
frames. This approach has several benefits over traditional object detection: it
is incredibly fast, lightweight and protects the privacy of its subjects.
We have trained and evaluated several neural network models tasked with
detecting and counting vehicles in various scenes and have achieved accuracies
above 90%. Finally, we take the first steps toward a decentralized solution
running entirely on embedded devices. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/8962789
- author
- Hjelm, Sebastian LU and Gustafsson, Mattias LU
- supervisor
-
- Jörn Janneck LU
- organization
- course
- EDAM05 20181
- year
- 2018
- type
- H3 - Professional qualifications (4 Years - )
- subject
- keywords
- Machine learning, Neural networks, Vehicle detection, CNN, Metadata, Bitrate, QP, Fast, Privacy
- publication/series
- LU-CS-EX 2018-13
- report number
- LU-CS-EX 2018-13
- ISSN
- 1650-2884
- language
- English
- id
- 8962789
- date added to LUP
- 2018-12-19 13:38:31
- date last changed
- 2018-12-19 13:38:31
@misc{8962789, abstract = {{The current field of object detection and image recognition is huge but not without complications. Processing large amounts of high resolution videos needs powerful hardware and also risks breaching the privacy of those who are recorded. In times of increasing demand for decentralized solutions and stricter privacy protection regulations being put in place a new approach is needed. We present an alternative to traditional object detection in video where we analyze changes to its metadata over time rather than the content of the video frames. This approach has several benefits over traditional object detection: it is incredibly fast, lightweight and protects the privacy of its subjects. We have trained and evaluated several neural network models tasked with detecting and counting vehicles in various scenes and have achieved accuracies above 90%. Finally, we take the first steps toward a decentralized solution running entirely on embedded devices.}}, author = {{Hjelm, Sebastian and Gustafsson, Mattias}}, issn = {{1650-2884}}, language = {{eng}}, note = {{Student Paper}}, series = {{LU-CS-EX 2018-13}}, title = {{Vehicle Counting using Video Metadata}}, year = {{2018}}, }