報(bào)告題目：Local Image Descriptors and Its Applications
報(bào)告人：楊學(xué)東 教授 博導(dǎo)
單位：加拿大里賈那大學(xué)計(jì)算機(jī)科學(xué)系
報(bào)告時(shí)間：2013年5月31日下午14:30-16:00
報(bào)告地點(diǎn)：逸夫樓408A會(huì)議室
報(bào)告摘要：
In pattern recognition and 3D computer vision, corresponding points between two images at low level are often the basis for object matching or identification at high level. Finding corresponding points from two images is usually based on the local image content surrounding the points under consideration. Early methods use simple template matching which may be computationally time consuming and sensitive to noise, and variation in illumination and view direction. Local image descriptor is a vector contains features derived from an image patch centered at a given point. It aims to provide translation, rotation and/or scaling invariant features for fast and reliable matching. It has been studied for about 14 years by researchers. This talk gives an overview of local image descriptors and some sample applications.
楊學(xué)東教授學(xué)習(xí)工作經(jīng)歷：
楊學(xué)東博士現(xiàn)任加拿大里賈那大學(xué)計(jì)算機(jī)科學(xué)系教授，系主任。他1982年畢業(yè)于太陽集團(tuán)tyc5997。1982-1988年期間在美國紐約大學(xué)（NYU）克朗數(shù)學(xué)研究所學(xué)習(xí)，獲得碩士及博士學(xué)位。1988-1990期間在康乃爾大學(xué)客座研究。1990至今在加拿大里賈那大學(xué)任教。
太陽集團(tuán)tyc5997
TOP
