Generic visual categorization method and system
First Claim
Patent Images
1. A method for assigning one of a plurality of classes to an input image, comprising:
- identifying a plurality of key-patches in the input image;
computing a feature vector for each of (at least ones of) the plurality of key-patches;
computing a histogram for each of (at least ones of) the plurality of classes using the plurality of feature vectors computed;
assigning at least one of the plurality of classes to the input image using the plurality of histograms computed as input to a classifier.
6 Assignments
0 Petitions
Accused Products
Abstract
Generic visual categorization methods complement a general vocabulary with adapted vocabularies that are class specific. Images to be categorized are characterized within different categories through a histogram indicating whether the image is better described by the general vocabulary or the class-specific adapted vocabulary.
149 Citations
20 Claims
-
1. A method for assigning one of a plurality of classes to an input image, comprising:
-
identifying a plurality of key-patches in the input image;
computing a feature vector for each of (at least ones of) the plurality of key-patches;
computing a histogram for each of (at least ones of) the plurality of classes using the plurality of feature vectors computed;
assigning at least one of the plurality of classes to the input image using the plurality of histograms computed as input to a classifier. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8)
-
-
9. An apparatus for assigning one of a plurality of classes to an input image, comprising:
-
a key-patch detector identifying a plurality of key-patches in the input image;
a feature description module for computing a feature vector for each of (at least ones of) the plurality of key-patches;
a multi-histogram computation module computing a histogram for each of (at least ones of) the plurality of classes using the plurality of feature vectors computed;
a classifier for assigning at least one of the plurality of classes to the input image using the plurality of histograms computed as input to the classifier. - View Dependent Claims (10, 11, 12, 13, 14, 15, 16)
-
-
17. A method for training a classifier, comprising:
-
identifying key-patches in images of a plurality of class training sets;
computing feature vectors for the identified key-patches;
computing a general vocabulary by clustering the computed feature vectors;
for each of a plurality of classes, computing an adapted vocabulary using the general vocabulary;
computing a histogram for each of the plurality of classes;
training the classifier using the histograms for each of the plurality of classes. - View Dependent Claims (18, 19, 20)
-
Specification