| Abstract | This paper presents a method for detecting categories of objects in real-worldimages. Given training images of an object category, our goal is to recognize
and localize instances of those objects in a candidate image.
The main contribution of this work is a novel structure of the shape code-
book for object detection. A shape codebook entry consists of two compo-
nents: a shape codeword and a group of associated vectors that specify the
object centroids. Like their counterpart in language, the shape codewords are
simple and generic such that they can be easily extracted from most object
categories. The associated vectors store the geometrical relationships be-
tween the shape codewords, which specify the characteristics of a particular
object category. Thus they can be considered as the “grammar” of the shape
codebook.
In this paper, we use Triple-Adjacent-Segments (TAS) extracted from im-
age edges as the shape codewords. Object detection is performed in a prob-
abilistic voting framework. Experimental results on public datasets show
performance similiar to the state-of-the-art, yet our method has significantly
lower complexity and requires considerably less supervision in the training
(We only need bounding boxes for a few training samples, do not need fig-
ure/ground segmentation and do not need a validation dataset).
|
