October 19, 2016
2:00 pm - 5:30 pm
Hall C

Track: General
Type: Posters
Level: All

We introduce a new local feature descriptor by exploring the pseudo-Wigner distribution of images and the Local Binary Patterns. It utilizes frequency domain smoothing, bag-of-words model and spatial pyramid representations to improve classification performance for scene images. Experiments with three datasets show that the WLBP outperforms traditional LBP and some other popular descriptors.

Speaker(s)