题目：Unifying Discriminative Visual Codebook Generation with Classifier Train
ing for Object Category Recognition 时间：12月24日 14:30-15:30
地点：蒙民伟楼404会议室 报告人：
Rong Jin
Associate Professor
Department of Computer Science, Michigan State University, USA
摘要：
The idea of representing images using a bag of visual words is currently popul
ar in object category recognition. Since this representation is typically cons
tructed using unsupervised clustering, the resulting visual words may not
capture the desired information. Recent work has explored the construction of
discriminative visual codebooks that explicitly consider object category infor
mation. However, since the codebook generation process is still disconnected f
rom that of classifier training, the set of resulting visual
words, while individually discriminative, may not be those best suited for the
classifier. In this talk, I will discuss proposes a novel optimization framew
ork that unifies codebook generation with classifier training. In this new fra
mework, each image feature is encoded by a sequence of "visual bits" optimized
for each category. An image, which can contain objects from multiple categori
es, is represented using aggregates of visual bits for each category. Experime
nts compare our approach to standard clustering-based methods and with state-o
f-the-art discriminative visual codebook generation. The significant improveme
nts over previous techniques clearly demonstrate the value of unifying represe
ntation and classification into a single optimization framework. 简历：
Rong Jin is an Associate Professor in the Computer and Science Engineering
Department at Michigan State University since 2008. His research is focused on
statistical machine learning and its application to large-scale information m
anagement. He has published over a hundred conference and journal articles on
the related topics. Dr. Jin holds a B.A. in Engineering from Tianjin Universit
y, an M.S. in Physics from Beijing University, and an M.S. and Ph.D. in Comput
er Science from Carnegie Mellon University. He received the NSF Career Award i
n 2006. |
