PT Unknown
AU Joost Van de Weijer
   Fahad Shahbaz Khan
TI Fusing Color and Shape for Bag-of-Words Based Object Recognition
BT 4th Computational Color Imaging Workshop
PY 2013
BP 25
EP 34
VL 7786
DI 10.1007/978-3-642-36700-7_3
DE Object Recognition; color features; bag-of-words; image classification
AB In this article we provide an analysis of existing methods for the incorporation of color in bag-of-words based image representations. We propose a list of desired properties on which bases fusing methods can be compared. We discuss existing methods and indicate shortcomings of the two well-known fusing methods, namely early and late fusion. Several recent works have addressed these shortcomings by exploiting top-down information in the bag-of-words pipeline: color attention which is motivated from human vision, and Portmanteau vocabularies which are based on information theoretic compression of product vocabularies. We point out several remaining challenges in cue fusion and provide directions for future research.
ER