- author = "Michal Wiglasz and Lukas Sekanina",
- title = "Evolutionary Approximation of Gradient Orientation Module in {HOG-based} Human Detection System",
- booktitle = "2017 IEEE Global Conference on Signal and Information Processing GlobalSIP 2017",
- year = "2017",
- editor = "Z. Jane Wang and Costas Kotropoulos and Qionghai Dai",
- pages = "1300--1304",
- address = "Montreal",
- month = nov # " 14-16",
- publisher = "IEEE Signal Processing Society",
- keywords = "genetic algorithms, genetic programming, Cartesian genetic programming, functional approximation, Histogram of oriented gradients: Poster",
- isbn13 = "978-1-5090-5989-8",
- language = "english",
-
URL = "
http://www.fit.vutbr.cz/research/view_pub.php?id=11441",
-
URL = "
http://www.fit.vutbr.cz/~iwiglasz/pubs.php.en?id=11441&yfile=%2Fpub%2F11441%2F0001300.pdf",
-
DOI = "
10.1109/GlobalSIP.2017.8309171",
- size = "5 page",
- abstract = "The histogram of oriented gradients (HOG) feature extraction is a computer vision method widely used in embedded systems for detection of objects such as pedestrians. We used Cartesian genetic programming (CGP) to exploit the error resilience in the HOG algorithm. We evolved new approximate implementations of the arctangent function, which is typically employed to compute the gradient orientations. When the best evolved approximations are integrated into the SW implementation of the HOG algorithm, not only the execution time, but also the classification accuracy was improved in comparison with the accurate implementation and the state-of-the art approximate implementations.",
-
notes = "Replaced by \cite{Wiglasz:2018:ieeeCompIntl}
MIT and INRIA pedestrian datasets. Lena image. SVM. Intel Xeon (with hardware support for arctan)
Approximate 8-bit arctan evolved by CGP \cite{Sekanina:2011:CGP.ch6} (might replace CORDIC iterative algorithm). Fitness based on hits.
https://2017.ieeeglobalsip.org/Papers/PublicSessionIndex3.asp?Sessionid=1063 Also known as \cite{8309171}",
