Image annotation using metric learning in semantic neighbourhoods

A Multi-View Embedding Space for Modeling Internet Images, Tags, and Their Semantics

Abstract: This paper investigates the problem of modeling Internet images and associated text or tags for tasks such as image-to-image search, tag-to-image search, and image-to-tag search (image annotation). We start with canonical correlation analysis (CCA), a popular and successful approach for mapping visual and textual features to the same latent space, and incorporate a third view capturing high-level image semantics, represented either by a single category or multiple non-mutually-exclusive concepts. We present two ways to train the three-view embedding: supervised, with the third view coming from ground-truth labels or search keywords; and unsupervised, with semantic themes automatically obtained by clustering the tags. To ensure high accuracy for retrieval tasks while keeping the learning process scalable, we combine multiple strong visual features and use explicit nonlinear kernel mappings to efficiently approximate kernel CCA. To perform retrieval, we use a specially designed similarity function in the embedded space, which substantially outperforms the Euclidean distance. The resulting system produces compelling qualitative results and outperforms a number of two-view baselines on retrieval tasks on three large-scale Internet image datasets.

NMF-KNN: Image Annotation Using Weighted Multi-view Non-negative Matrix Factorization

Abstract: The real world image databases such as Flickr are characterized by continuous addition of new images. The recent approaches for image annotation, i.e. the problem of assigning tags to images, have two major drawbacks. First, either models are learned using the entire training data, or to handle the issue of dataset imbalance, tag-specific discriminative models are trained. Such models become obsolete and require relearning when new images and tags are added to database. Second, the task of feature-fusion is typically dealt using ad-hoc approaches. In this paper, we present a weighted extension of Multi-view Non-negative Matrix Factorization (NMF) to address the aforementioned drawbacks. The key idea is to learn query-specific generative model on the features of nearest-neighbors and tags using the proposed NMF-KNN approach which imposes consensus constraint on the coefficient matrices across different features. This results in coefficient vectors across features to be consistent and, thus, naturally solves the problem of feature fusion, while the weight matrices introduced in the proposed formulation alleviate the issue of dataset imbalance. Furthermore, our approach, being query-specific, is unaffected by addition of images and tags in a database. We tested our method on two datasets used for evaluation of image annotation and obtained competitive results.

Automatic Image Annotation using Deep Learning Representations

Abstract: We propose simple and effective models for the image annotation that make use of Convolutional Neural Network (CNN) features extracted from an image and word embedding vectors to represent their associated tags. Our first set of models is based on the Canonical Correlation Analysis (CCA) framework that helps in modeling both views - visual features (CNN feature) and textual features (word embedding vectors) of the data. Results on all three variants of the CCA models, namely linear CCA, kernel CCA and CCA with k-nearest neighbor (CCA-KNN) clustering, are reported. The best results are obtained using CCA-KNN which outperforms previous results on the Corel-5k and the ESP-Game datasets and achieves comparable results on the IAPRTC-12 dataset. In our experiments we evaluate CNN features in the existing models which bring out the advantages of it over dozens of handcrafted features. We also demonstrate that word embedding vectors perform better than binary vectors as a representation of the tags associated with an image. In addition we compare the CCA model to a simple CNN based linear regression model, which allows the CNN layers to be trained using back-propagation.

Love Thy Neighbors: Image Annotation by Exploiting Image Metadata

Abstract: Some images that are difficult to recognize on their own may become more clear in the context of a neighborhood of related images with similar social-network metadata. We build on this intuition to improve multilabel image annotation. Our model uses image metadata nonparametrically to generate neighborhoods of related images using Jaccard similarities, then uses a deep neural network to blend visual information from the image and its neighbors. Prior work typically models image metadata parametrically, in contrast, our nonparametric treatment allows our model to perform well even when the vocabulary of metadata changes between training and testing. We perform comprehensive experiments on the NUS-WIDE dataset, where we show that our model outperforms state-of-the-art methods for multilabel image annotation even when our model is forced to generalize to new types of metadata.

Multi-Label Dictionary Learning for Image Annotation

Abstract: Image annotation has attracted a lot of research interest, and multi-label learning is an effective technique for image annotation. How to effectively exploit the underlying correlation among labels is a crucial task for multi-label learning. Most existing multi-label learning methods exploit the label correlation only in the output label space, leaving the connection between the label and the features of images untouched. Although, recently some methods attempt toward exploiting the label correlation in the input feature space by using the label information, they cannot effectively conduct the learning process in both the spaces simultaneously, and there still exists much room for improvement. In this paper, we propose a novel multi-label learning approach, named multi-label dictionary learning (MLDL) with label consistency regularization and partial-identical label embedding MLDL, which conducts MLDL and partial-identical label embedding simultaneously. In the input feature space, we incorporate the dictionary learning technique into multi-label learning and design the label consistency regularization term to learn the better representation of features. In the output label space, we design the partial-identical label embedding, in which the samples with exactly same label set can cluster together, and the samples with partial-identical label sets can collaboratively represent each other. Experimental results on the three widely used image datasets, including Corel 5K, IAPR TC12, and ESP Game, demonstrate the effectiveness of the proposed approach.

Distance Metric Learning for Large Margin Nearest Neighbor Classification

Abstract: We show how to learn a Mahanalobis distance metric for k-nearest neighbor (kNN) classification by semidefinite programming. The metric is trained with the goal that the k-nearest neighbors always belong to the same class while examples from different classes are separated by a large margin. On seven data sets of varying size and difficulty, we find that metrics trained in this way lead to significant improvements in kNN classification—for example, achieving a test error rate of 1.3% on the MNIST handwritten digits. As in support vector machines (SVMs), the learning problem reduces to a convex optimization based on the hinge loss. Unlike learning in SVMs, however, our framework requires no modification or extension for problems in multiway (as opposed to binary) classification.

Distance Metric Learning for Large Margin Nearest Neighbor Classification

Abstract: The accuracy of k-nearest neighbor (kNN) classification depends significantly on the metric used to compute distances between different examples. In this paper, we show how to learn a Mahalanobis distance metric for kNN classification from labeled examples. The Mahalanobis metric can equivalently be viewed as a global linear transformation of the input space that precedes kNN classification using Euclidean distances. In our approach, the metric is trained with the goal that the k-nearest neighbors always belong to the same class while examples from different classes are separated by a large margin. As in support vector machines (SVMs), the margin criterion leads to a convex optimization based on the hinge loss. Unlike learning in SVMs, however, our approach requires no modification or extension for problems in multiway (as opposed to binary) classification. In our framework, the Mahalanobis distance metric is obtained as the solution to a semidefinite program. On several data sets of varying size and difficulty, we find that metrics trained in this way lead to significant improvements in kNN classification. Sometimes these results can be further improved by clustering the training examples and learning an individual metric within each cluster. We show how to learn and combine these local metrics in a globally integrated manner.

Labeling images with a computer game

Abstract: We introduce a new interactive system: a game that is fun and can be used to create valuable output. When people play the game they help determine the contents of images by providing meaningful labels for them. If the game is played as much as popular online games, we estimate that most images on the Web can be labeled in a few months. Having proper labels associated with each image on the Web would allow for more accurate image search, improve the accessibility of sites (by providing descriptions of images to visually impaired individuals), and help users block inappropriate images. Our system makes a significant contribution because of its valuable output and because of the way it addresses the image-labeling problem. Rather than using computer vision techniques, which don't work well enough, we encourage people to do the work by taking advantage of their desire to be entertained.

Pegasos: primal estimated sub-gradient solver for SVM

Abstract: We describe and analyze a simple and effective stochastic sub-gradient descent algorithm for solving the optimization problem cast by Support Vector Machines (SVM). We prove that the number of iterations required to obtain a solution of accuracy $${\epsilon}$$ is $${\tilde{O}(1 / \epsilon)}$$, where each iteration operates on a single training example. In contrast, previous analyses of stochastic gradient descent methods for SVMs require $${\Omega(1 / \epsilon^2)}$$ iterations. As in previously devised SVM solvers, the number of iterations also scales linearly with 1/λ, where λ is the regularization parameter of SVM. For a linear kernel, the total run-time of our method is $${\tilde{O}(d/(\lambda \epsilon))}$$, where d is a bound on the number of non-zero features in each example. Since the run-time does not depend directly on the size of the training set, the resulting algorithm is especially suited for learning from large datasets. Our approach also extends to non-linear kernels while working solely on the primal objective function, though in this case the runtime does depend linearly on the training set size. Our algorithm is particularly well suited for large text classification problems, where we demonstrate an order-of-magnitude speedup over previous SVM learning methods.

Object Recognition as Machine Translation: Learning a Lexicon for a Fixed Image Vocabulary

Abstract: We describe a model of object recognition as machine translation. In this model, recognition is a process of annotating image regions with words. Firstly, images are segmented into regions, which are classified into region types using a variety of features. A mapping between region types and keywords supplied with the images, is then learned, using a method based around EM. This process is analogous with learning a lexicon from an aligned bitext. For the implementation we describe, these words are nouns taken from a large vocabulary. On a large test set, the method can predict numerous words with high accuracy. Simple methods identify words that cannot be predicted well. We show how to cluster words that individually are difficult to predict into clusters that can be predicted well -- for example, we cannot predict the distinction between train and locomotive using the current set of features, but we can predict the underlying concept. The method is trained on a substantial collection of images. Extensive experimental results illustrate the strengths and weaknesses of the approach.