about me

Ground Plane Classification using edge based segmentation

Post replyCan't edit post
Watch topicCan't delete post
Monday, April 23 2007
Download the document below.
Due to license restrictions, only a single downsampled cyclorama has been supplied with the source code.
Abstract: Feedforward artificial neural networks, popular in image processing and understanding, are elegant computational machines focused on approximating nonlinear functions. This paper takes a closer look at trying to understand visual data by making optimal use of neural networks. Provided a set of input-output pairs derived from 360 panoramic photographs established using a strict series of steps the automatic classification and extraction of the ground plane in the images is tested and scored against human results.
gpc.pdf (1.1 MB)
gpc_manual.pdf (49.43 KB) (1.02 MB)
Written in TeX and compiled by MiKTeX
Programmed for Java 1.6 and up
Actions:Post replyWatch topic
No comments.Post reply
Please contact me prior to distributing any files.
Members have extra privileges. You can login or register.
© 2005–2015 P. F. Lammertsma
No members currently online; 593 hits today by 157 unique users