We have witnessed great interest and a wealth of promise in content-based image retrieval as an emerging technology. While the last decade laid foundation to such promise, it also paved the way for a large number of new techniques and systems, got many new people involved, and triggered stronger association of weakly related fields. In this article, we survey almost 300 key theoretical and empirical contributions in the current decade related to image retrieval and automatic image annotation, and in the process discuss the spawning of related subfields. We also discuss significant challenges involved in the adaptation of existing image retrieval techniques to build systems that can be useful in the real world. In retrospect of what has been achieved so far, we also conjecture what the future may hold for image retrieval research.

We propose a content-based soft annotation (CBSA) procedure for providing images with semantical labels. The annotation procedure starts with labeling a small set of training images, each with one single semantical label (e.g., forest, animal, or sky). An ensemble of binary classifiers is then trained for predicting label membership for images. The trained ensemble is applied to each individual image to give the image multiple soft labels, and each label is associated with a label membership factor. To select a base binary-classifier for CBSA, we experiment with two learning methods, Support Vector Machines (SVMs) and Bayes Point Machines (BPMs, and compare their class-prediction accuracy. Our empirical study on a 116-category 25K-image set shows that the BPM-based ensemble provides better annotation quality than the SVM-based ensemble for supporting multimodal image retrievals. Keywords: Bayes Point Machines, Support Vector Machines, image annotation, multimodal image retrieval.

Abstract. We extend the constellation model to include heterogeneous parts which may represent either the appearance or the geometry of a region of the object. The parts and their spatial configuration are learnt simultaneously and automatically, without supervision, from cluttered images. We describe how this model can be employed for ranking the output of an image search engine when searching for object categories. It is shown that visual consistencies in the output images can be identified, and then used to rank the images according to their closeness to the visual object category. Although the proportion of good images may be small, the algorithm is designed to be robust and is capable of learning in either a totally unsupervised manner, or with a very limited amount of supervision. We demonstrate the method on image sets returned by Google’s image search for a number of object categories including bottles, camels, cars, horses, tigers and zebras.

As current methods for content-based retrieval are incapable of capturing the semantics of images, we experiment with using spectral methods to infer a semantic space from user’s relevance feedback, so that our system will gradually improve its retrieval performance through accumulated user interactions. In addition to the long-term learning process, we also model the traditional approaches to query refinement using relevance feedback as a short-term learning process. The proposed shortand long-term learning frameworks have been integrated into an image retrieval system. Experimental results on a large collection of images have shown the effectiveness and robustness of our proposed algorithms.

In this paper, we propose a novel transductive learning framework named manifold-ranking based image retrieval (MRBIR). Given a query image, MRBIR first makes use of a manifold ranking algorithm to explore the relationship among all the data points in the feature space, and then measures relevance between the query and all the images in the database accordingly, which is different from traditional similarity metrics based on pair-wise distance. In relevance feedback, if only positive examples are available, they are added to the query set to improve the retrieval result; if examples of both labels can be obtained, MRBIR discriminately spreads the ranking scores of positive and negative examples, considering the asymmetry between these two types of images. Furthermore, three active learning methods are incorporated into MRBIR, which select images in each round of relevance feedback according to different principles, aiming to maximally improve the ranking result. Experimental results on a general-purpose image database show that MRBIR attains a significant improvement over existing systems from all aspects.

Abstract. Support Vector Machines (SVM) have been extensively studied and have shown remarkable success in many applications. However the success of SVM is very limited when it is applied to the problem of learning from imbalanced datasets in which negative instances heavily outnumber the positive instances (e.g. in gene profiling and detecting credit card fraud). This paper discusses the factors behind this failure and explains why the common strategy of undersampling the training data may not be the best choice for SVM. We then propose an algorithm for overcoming these problems which is based on a variant of the SMOTE algorithm by Chawla et al, combined with Veropoulos et al’s different error costs algorithm. We compare the performance of our algorithm against these two algorithms, along with undersampling and regular SVM and show that our algorithm outperforms all of them. 1

The paper is concerned with two-class active learning. While the common approach for collecting data in active learning is to select samples close to the classification boundary, better performance can be achieved by taking into account the prior data distribution. The main contribution of the paper is a formal framework that incorporates clustering into active learning. The algorithm first constructs a classifier on the set of the cluster representatives, and then propagates the classification decision to the other samples via a local noise model. The proposed model allows to select the most representative samples as well as to avoid repeatedly labeling samples in the same cluster. During the active learning process, the clustering is adjusted using the coarse-to-fine strategy in order to balance between the advantage of large clusters and the accuracy of the data representation. The results of experiments in image databases show a better performance of our algorithm compared to the current methods.

In this paper we unify two supposedly distinct tasks in multimedia retrieval. One task involves answering queries with a few examples. The other involves learning models for semantic concepts, also with a few examples. In our view these two tasks are identical with the only differentiation being the number of examples that are available for training. Once we adopt this unified view, we then apply identical techniques for solving both problems and evaluate the performance using the NIST TRECVID benchmark evaluation data [15]. We propose a combination hypothesis of two complementary classes of techniques, a nearest neighbor model using only positive examples and a discriminative support vector machine model using both positive and negative examples. In case of queries, where negative examples are rarely provided to seed the search, we create pseudo-negative samples. We then combine the ranked lists generated by evaluating the test database using both methods, to create a final ranked list of retrieved multimedia items. We evaluate this approach for rare concept and query topic modeling using the NIST TRECVID video corpus. In both tasks we find that applying the combination hypothesis across both modeling techniques and a variety of features results in enhanced performance over any of the baseline models, as well as in improved robustness with respect to training examples and visual features. In particular, we observe an improvement of 6 % for rare concept detection and 17 % for the search task.

In a typical content-based image retrieval (CBIR) system, query results are a set of images sorted by feature similarities with respect to the query. However, images with high feature similarities to the query may be very di#erent from the query in terms of semantics. This discrepancy between low-level features and high-level concepts is known as the semantic gap. This paper introduces a novel image retrieval scheme, CLUster-based rEtrieval of images by unsupervised learning (CLUE), which attempts to tackle the semantic gap problem based on a hypothesis that images of the same semantics are similar in a way, images of di#erent semantics are di#erent in their own ways. CLUE attempts to capture semantic concepts by learning the way that images of the same semantics are similar and retrieving image clusters instead of a set of ordered images. Clustering in CLUE is dynamic. In particular, clusters formed depend on which images are retrieved in response to the query. Therefore, the clusters give the algorithm as well as the users semantic relevant clues as to where to navigate. CLUE is a general approach that can be combined with any real-valued symmetric similarity measure (metric or nonmetric). Thus it may be embedded in many current CBIR systems. An experimental image retrieval system using CLUE has been implemented. The performance of the system is evaluated on a database of about 60, 000 images from COREL. Empirical results demonstrate improved performance compared with a typical CBIR system using the same image similarity measure. In addition, preliminary results on images returned by Google's Image Search reveal the potential of applying CLUE to real world image data and integrating CLUE as a part of the interface for keyword-based image retrieval systems.

We consider the problem of learning a mapping function from low-level feature space to high-level semantic space. Under the assumption that the data lie on a submanifold embedded in a high dimensional Euclidean space, we propose a relevance feedback scheme which is naturally conducted only on the image manifold in question rather than the total ambient space. While images are typically represented by feature vectors in R n, the natural distance is often different from the distance induced by the ambient space R n. The geodesic distances on manifold are used to measure the similarities between images. However, when the number of data points is small, it is hard to discover the intrinsic manifold structure. Based on user interactions in a relevance feedback driven query-by-example system, the intrinsic similarities between images can be accurately estimated. We then develop an algorithmic framework to approximate the optimal mapping function by a Radial Basis Function (RBF) neural network. The semantics of a new image can be inferred by the RBF neural network. Experimental results show that our approach is effective in improving the performance of content-based image retrieval systems.