Retrieval system and method leveraging category-level labels
First Claim
Patent Images
1. A retrieval method comprising:
- learning a projection for embedding an original image representation in an embedding space, the original image representation being based on features extracted from the image, the projection being learned from category-labeled training data to optimize a classification rate on the training data, the learning of the projection including, for a plurality of iterations;
selecting a sample from the training data;
embedding the sample with a current projection;
scoring the embedded sample with current first and second classifiers, the first classifier corresponding to a category of the label of the sample, the second classifier corresponding to a different category, selected from a set of categories;
updated the current projection and at least one of the current first and second classifier for iterations where the second classifier generates a higher score than the first classifier, the updated projection serving as the current projection for a subsequent iteration, each of the updated classifiers serving as the current classifier for the respective category for a subsequent iteration; and
storing one of the updated projections as the learned projection; and
with a processor, for each of plurality of database images, computing a comparison measure between a query image and the database image, the comparison measure being computed in the embedding space, respective original image representations of the query image and the database image being embedded in the embedding space with the projection; and
providing for retrieving at least one of the database images based on the comparison.
1 Assignment
0 Petitions
Accused Products
Abstract
An instance-level retrieval method and system are provided. A representation of a query image is embedded in a multi-dimensional space using a learned projection. The projection is learned using category-labeled training data to optimize a classification rate on the training data. The joint learning of the projection and the classifiers improves the computation of similarity/distance between images by embedding them in a subspace where the similarity computation outputs more accurate results. An input query image can thus be used to retrieve similar instances in a database by computing the comparison measure in the embedding space.
56 Citations
27 Claims
-
1. A retrieval method comprising:
-
learning a projection for embedding an original image representation in an embedding space, the original image representation being based on features extracted from the image, the projection being learned from category-labeled training data to optimize a classification rate on the training data, the learning of the projection including, for a plurality of iterations; selecting a sample from the training data; embedding the sample with a current projection; scoring the embedded sample with current first and second classifiers, the first classifier corresponding to a category of the label of the sample, the second classifier corresponding to a different category, selected from a set of categories; updated the current projection and at least one of the current first and second classifier for iterations where the second classifier generates a higher score than the first classifier, the updated projection serving as the current projection for a subsequent iteration, each of the updated classifiers serving as the current classifier for the respective category for a subsequent iteration; and storing one of the updated projections as the learned projection; and with a processor, for each of plurality of database images, computing a comparison measure between a query image and the database image, the comparison measure being computed in the embedding space, respective original image representations of the query image and the database image being embedded in the embedding space with the projection; and providing for retrieving at least one of the database images based on the comparison. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18)
-
-
19. A retrieval method comprising:
-
with a processor, learning a projection for embedding an original image representation in an embedding space, the original image representation being based on features extracted from the image, the projection being learned from category-labeled training data to optimize a classification rate on the training data, the learning of the projection including optimizing an objective function of the form;
Σ
(q,y+,y−
)min{0,t−
s(q,y+)+s(q,y−
)}where t represents a predetermined threshold, q represents a sample, s(q,y+) represents a score of the sample on the classifier corresponding to its category and s(q,y−
) represents a score of the sample on a classifier not corresponding to its category; andfor each of plurality of data base images, computing a comparison measure between a query image and the database image, the comparison measure being computed in the embedding space, respective original image representations of the query image and the database image being embedded in the embedding space with the projection; and providing for retrieving at least one of the database images based on the comparison. - View Dependent Claims (20)
-
-
21. A retrieval system comprising:
-
memory which stores; a projection matrix for embedding image features in an embedding space, the projection matrix having been learned from category-labeled training data to optimize a classification rate on the training data, including, for a plurality of iterations; selecting a sample from the training data; embedding the sample with a current projection; scoring the embedded sample with current first and second classifiers, the first classifier corresponding to a category of the label of the sample, the second classifier corresponding to a different category, selected from a set of categories; updating the current projection and at least one of the current first and second classifiers for iterations where the second classifier generates a higher score than the first classifier, the updated projection serving as the current projection for a subsequent iteration, each of the updated classifiers serving as the current classifier for the respective category for a subsequent iteration; and storing an updated projection as the learned projection; and instructions for computing a comparison between a query image and a database image whose respective features are embedded in the embedding space with the projection matrix; and a processor in communication with the memory which implements the instructions. - View Dependent Claims (22, 23, 24)
-
-
25. A retrieval method comprising:
-
providing a feature-based representation and a category label for each of a set of training images, each of the category labels corresponding to a respective one of a set of categories; jointly learning a projection and set of classifiers based on the feature-based representations and category labels, the learning optimizing a classification of the training images by the set of classifiers in an embedding space into which the feature-based representations are embedded with the projection, the set of classifiers including a classifier for each of the categories; storing the projection for embedding a query image and database images into the embedding space; receiving a query image; and without using the learned set of classifiers, retrieving at least one the database images based on a computed comparison measure between the query image and the database images embedded in the embedding space with the learned projection. - View Dependent Claims (26, 27)
-
Specification