Tobias Gass

Jointly optimising relevance and diversity in image retrieval

In this paper we present a method to jointly optimise the relevance and the diversity of the results in image retrieval. Without considering diversity, image retrieval systems often mainly find a set of very similar results, so called near duplicates, which is often not the desired behaviour. From the user perspective, the ideal result consists of documents which are not only relevant but ideally also diverse. Most approaches addressing diversity in image or information retrieval use a two-step approach where in a first step a set of potentially relevant images is determined and in a second step these images are reranked to be diverse among the first positions. In contrast to these approaches, our method addresses the problem directly and jointly optimises the diversity and the relevance of the images in the retrieval ranking using techniques inspired by dynamic programming algorithms. We quantitatively evaluate our method on the ImageCLEF 2008 photo retrieval data and obtain results which outperform the state of the art. Additionally, we perform a qualitative evaluation on a new product search task and it is observed that the diverse results are more attractive to an average user.

Equivalence of Generative and Log-Linear Models

Conventional speech recognition systems are based on hidden Markov models (HMMs) with Gaussian mixture models (GHMMs). Discriminative log-linear models are an alternative modeling approach and have been investigated recently in speech recognition. GHMMs are directed models with constraints, e.g., positivity of variances and normalization of conditional probabilities, while log-linear models do not use such constraints. This paper compares the posterior form of typical generative models related to speech recognition with their log-linear model counterparts. The key result will be the derivation of the equivalence of these two different approaches under weak assumptions. In particular, we study Gaussian mixture models, part-of-speech bigram tagging models, and eventually, the GHMMs. This result unifies two important but fundamentally different modeling paradigms in speech recognition on the functional level. Furthermore, this paper will present comparative experimental results for various speech tasks of different complexity, including a digit string and large-vocabulary continuous speech recognition tasks.

Cloud-Based Evaluation of Anatomical Structure Segmentation and Landmark Detection Algorithms: VISCERAL Anatomy Benchmarks

Variations in the shape and appearance of anatomical structures in medical images are often relevant radiological signs of disease. Automatic tools can help automate parts of this manual process. A cloud-based evaluation framework is presented in this paper including results of benchmarking current state-of-the-art medical imaging algorithms for anatomical structure segmentation and landmark detection: the VISCERAL Anatomy benchmarks. The algorithms are implemented in virtual machines in the cloud where participants can only access the training data and can be run privately by the benchmark administrators to objectively compare their performance in an unseen common test set. Overall, 120 computed tomography and magnetic resonance patient volumes were manually annotated to create a standard Gold Corpus containing a total of 1295 structures and 1760 landmarks. Ten participants contributed with automatic algorithms for the organ segmentation task, and three for the landmark localization task. Different algorithms obtained the best scores in the four available imaging modalities and for subsets of anatomical structures. The annotation framework, resulting data set, evaluation setup, results and performance analysis from the three VISCERAL Anatomy benchmarks are presented in this article. Both the VISCERAL data set and Silver Corpus generated with the fusion of the participant algorithms on a larger set of non-manually-annotated medical images are available to the research community.

Isotropic Total Variation Regularization of Displacements in Parametric Image Registration.

Spatial regularization is essential in image registration, which is an ill-posed problem. Regularization can help to avoid both physically implausible displacement fields and local minima during optimization. Tikhonov regularization (squared

Simultaneous Segmentation and Multiresolution Nonrigid Atlas Registration

\ell _{2}

Image warping for face recognition: From local optimality towards global optimization

-norm) is unable to correctly represent non-smooth displacement fields, that can, for example, occur at sliding interfaces in the thorax and abdomen in image time-series during respiration. In this paper, isotropic Total Variation (TV) regularization is used to enable accurate registration near such interfaces. We further develop the TV-regularization for parametric displacement fields and provide an efficient numerical solution scheme using the Alternating Directions Method of Multipliers (ADMM). The proposed method was successfully applied to four clinical databases which capture breathing motion, including CT lung and MR liver images. It provided accurate registration results for the whole volume. A key strength of our proposed method is that it does not depend on organ masks that are conventionally required by many algorithms to avoid errors at sliding interfaces. Furthermore, our method is robust to parameter selection, allowing the use of the same parameters for all tested databases. The average target registration error (TRE) of our method is superior (10% to 40%) to other techniques in the literature. It provides precise motion quantification and sliding detection with sub-pixel accuracy on the publicly available breathing motion databases (mean TREs of 0.95 mm for DIR 4D CT, 0.96 mm for DIR COPDgene, 0.91 mm for POPI databases).

Latent Log-Linear Models for Handwritten Digit Classification

In this paper, a novel Markov random field (MRF)-based approach is presented for segmenting medical images while simultaneously registering an atlas nonrigidly. In the literature, both segmentation and registration have been studied extensively. For applications that involve both, such as segmentation via atlas-based registration, earlier studies proposed addressing these problems iteratively by feeding the output of each to initialize the other. This scheme, however, cannot guarantee an optimal solution for the combined task at hand, since these two individual problems are then treated separately. In this paper, we formulate simultaneous registration and segmentation (SRS) as a maximum a-posteriori (MAP) problem. We decompose the resulting probabilities such that the MAP inference can be done using MRFs. An efficient hierarchical implementation is employed, allowing coarse-to-fine registration while estimating segmentation at pixel level. The method is evaluated on two clinical data sets: 1) mandibular bone segmentation in 3D CT and 2) corpus callosum segmentation in 2D midsaggital slices of brain MRI. A video tracking example is also given. Our implementation allows us to directly compare the proposed method with the individual segmentation/registration and the iterative approach using the exact same potential functions. In a leave-one-out evaluation, SRS demonstrated more accurate results in terms of dice overlap and surface distance metrics for both data sets. We also show quantitatively that the SRS method is less sensitive to the errors in the registration as opposed to the iterative approach.

Computer algorithms for three-dimensional measurement of humeral anatomy: analysis of 140 paired humeri

This paper systematically analyzes the strengths and weaknesses of existing image warping algorithms on the tasks of face recognition. Image warping is used to cope with local and global image variability and in general is an NP-complete problem. Although many approximations have recently been proposed, neither thorough comparison, nor systematic analysis of methods in a common scheme has been done so far. We follow the bottom-up approach and analyze the methods with increasing degree of image structure preserved during optimization. We evaluate the presented warping approaches on four challenging face recognition tasks in highly variable domains. Our findings indicate that preserving maximum dependencies between neighboring pixels by imposing strong geometrical constraints leads to the best recognition results while making optimization efficient.

Warp that smile on your face: Optimal and smooth deformations for face recognition

We present latent log-linear models, an extension of log-linear models incorporating latent variables, and we propose two applications thereof: log-linear mixture models and image deformation-aware log-linear models. The resulting models are fully discriminative, can be trained efficiently, and the model complexity can be controlled. Log-linear mixture models offer additional flexibility within the log-linear modeling framework. Unlike previous approaches, the image deformation-aware model directly considers image deformations and allows for a discriminative training of the deformation parameters. Both are trained using alternating optimization. For certain variants, convergence to a stationary point is guaranteed and, in practice, even variants without this guarantee converge and find models that perform well. We tune the methods on the USPS data set and evaluate on the MNIST data set, demonstrating the generalization capabilities of our proposed models. Our models, although using significantly fewer parameters, are able to obtain competitive results with models proposed in the literature.

Multi-atlas Segmentation and Landmark Localization in Images with Large Field of View

BACKGROUND In the presence of severe osteoarthritis, osteonecrosis, or proximal humeral fracture, the contralateral humerus may serve as a template for the 3-dimensional (3D) preoperative planning of reconstructive surgery. The purpose of this study was to develop algorithms for performing 3D measurements of the humeral anatomy and further to assess side-to-side (bilateral) differences in humeral head retrotorsion, humeral head inclination, humeral length, and humeral head radius and height. METHODS The 3D models of 140 paired humeri (70 cadavers) were extracted from computed tomographic data. Geometric characteristics quantifying the humeral anatomy in 3D were determined in a semiautomatic fashion using the developed computer algorithms. The results between the sides were compared for evaluating bilateral differences. RESULTS The mean bilateral difference of the humeral retrotorsion angle was 6.7° (standard deviation [SD], 5.7°; range, -15.1° to 24.0°; P = .063); the mean side difference of the humeral head inclination angle was 2.3° (SD, 1.8°; range, -5.1° to 8.4°; P = .12). The side difference in humeral length (mean, 2.9 mm; SD, 2.5 mm; range, -8.7 mm to 10.1 mm; P = .04) was significant. The mean side difference in the head sphere radius was 0.5 mm (SD, 0.6 mm; range, -3.2 mm to 2.2 mm; P = .76), and the mean side difference in humeral head height was 0.8 mm (SD, 0.6 mm; range, -2.4 mm to 2.4 mm; P = .44). CONCLUSIONS The contralateral anatomy may serve as a reliable reconstruction template for humeral length, humeral head radius, and humeral head height if it is analyzed with 3D algorithms. In contrast, determining humeral head retrotorsion and humeral head inclination from the contralateral anatomy may be more prone to error.