Dissimilarity Space Research Articles

Accelerating time series similarity search under Move-Split-Merge distance via dissimilarity space embedding

The ubiquity sensory devices across scientific and industrial settings have made time-series data mining an important research topic in the IoT domain. Among all time-series mining techniques, the similarity search is considered the most essential and noteworthy because it is frequently employed as a fundamental subroutine for other procedures. The Dynamic Time Warping (DTW) was once considered the best similarity measure in time-series data mining community; however, two recent studies have demonstrated that the Move-Split-Merge (MSM) performs better than DTW in both supervised and unsupervised settings. Computing MSM between two time-series incurs costly computation because MSM has quadratic time complexity. Regrettably, there is little research on accelerating time-series similarity search under MSM. This paper provides an efficient method named DSEMSM for speeding up this procedure. DSEMSM utilizes dissimilarity space embedding to transform time-series into low-dimensional vectors. The filter-and-verification framework is subsequently employed to reduce the large number of actual MSM computations and to identify the final search results from the candidate set. The furthest prototype selection strategy and lower bound based technique are applied in DSEMSM to improve its retrieval accuracy and efficiency, respectively. We validate the effectiveness of the proposed method on a large number of datasets. The experimental results indicate that, in comparison to its relevant competitors, DSEMSM achieves a more stable speedup rate and higher retrieval accuracy.

Visual Guidance for User Placement in Avatar-Mediated Telepresence Between Dissimilar Spaces.

Rapid advances in technology gradually realize immersive mixed-reality (MR) telepresence between distant spaces. This paper presents a novel visual guidance system for avatar-mediated telepresence, directing users to optimal placements that facilitate the clear transfer of gaze and pointing contexts through remote avatars in dissimilar spaces, where the spatial relationship between the remote avatar and the interaction targets may differ from that of the local user. Representing the spatial relationship between the user/avatar and interaction targets with angle-based interaction features, we assign recommendation scores of sampled local placements as their maximum feature similarity with remote placements. These scores are visualized as color-coded 2D sectors to inform the users of better placements for interaction with selected targets. In addition, virtual objects of the remote space are overlapped with the local space for the user to better understand the recommendations. We examine whether the proposed score measure agrees with the actual user perception of the partner's interaction context and find a score threshold for recommendation through user experiments in virtual reality (VR). A subsequent user study in VR investigates the effectiveness and perceptual overload of different combinations of visualizations. Finally, we conduct a user study in an MR telepresence scenario to evaluate the effectiveness of our method in real-world applications.

Open zero-shot learning via asymmetric VAE with dissimilarity space

Generalized Zero-Shot Learning (GZSL) aims to classify samples from seen and unseen classes using class-level features. Although existing GZSL methods have achieved remarkable progress, GZSL assumes a closed-set scenario, where both the seen and unseen classes are pre-defined. This assumption does not satisfy the complexity and variability of real-world scenarios. To address this, a new paradigm called Open Zero-Shot Learning (OZSL) has emerged, which aims to identify seen and unseen classes while rejecting the unknown ones where no semantic and visual features are available. Due to the absence of semantic features for unknown classes, the conventional methods adopted in GZSL are no longer applicable. Our research discovered that dissimilarities within the semantic space could be transferred to the visual space, achieving a shared dissimilarity space. Building upon this idea, we propose an asymmetric Variational Autoencoder (VAE) architecture. The encoders compute the dissimilarities within features and learn the latent distribution of dissimilarities, while the decoders synthesize features for unseen and unknown classes conditioned on original features and dissimilarities. We evaluate the proposed method on benchmark datasets, including AWA, CUB, SUN, and FLO, and achieve state-of-the-art OZSL performance.

A People’s Sea: Palestine and Popular Thalassopolitics in the Mediterranean Sea

This article draws on fieldwork, interviews, and archival research on the maritime campaign of the International Palestine Solidarity Movement to challenge the embargo on Gaza. I make three main arguments: first, in addressing the Mediterranean Sea as a platform for international solidarity, the Ships to Gaza inadvertently contrived a novel and largely understudied instance of popular politics at sea, which I coin ‘popular thalassopolitics’. I theorize this emergence in its own right and not solely within state-centric analyses or as a mere expansion of the terrestrial plane. This article urges us to read the sea through the lens of internationalist solidarity and the popular politics it invokes. Second, the popular thalassopolitics of the Ships charts and navigates an ‘insurgent terrain’, creatively assembling the well-established activist practice of solidarity vis-à-vis a decades-long indigenous struggle in Palestine with an emerging perception of the sea as a common and shared but dissimilar space for action. Thirdly, I argue that this history of the popular thalassopolitics of the Ships allows us to reread the scholarship on refugee rescue boats and civil activism at sea. This reading inadvertently challenges a Eurocentric bias in the emerging politicization of the sea as a humanitarianism space in which western saviors come to the aid of agentless refugees. La mer des peuples : La Palestine et la thalassopolitique populaire en mer Méditerranée

A novel cancelable finger vein templates based on LDM and RetinexGan

In this paper, we propose a new biometric template protection scheme, which can deal with the finger vein biometric security threats, through using the LDM and RetinexGAN model. The RetinexGAN model is mainly used to handle the illumination and low contrast problems effectively, while efficiently extracting discriminative features from the finger vein images. The projection of extracted features into dissimilarity space is done using Local Dissimilarity Map (LDM). LDM is an efficient way for finger vein features representation, which investigates the relationships and correlation inter and intra classes, while effectively coming up with the accidental shifts/rotations caused by the arbitrary position of the finger during image acquisition. The proposed approach is successfully evaluated in terms of non-invertibility, non-linkability, revocability and performances. Experimental results and comparison analysis with the state of arts methods confirm that the proposed framework can achieve promising results.

Unsupervised domain adaptation pedestrian re-identification based on an improved dissimilarity space

Pedestrian re-identification is a key and challenging research topic in intelligent security applications. Because of the need for large-scale manual labeling, pedestrian re-identification methods based on supervised learning cannot be widely used in practical applications. Research on unsupervised pedestrian re-identification has therefore become a hot topic. The DMMD model published in ECCV2020 successfully applied the dissimilarity space to unsupervised pedestrian re-identification. Based on the analysis of the deficiencies of DMMD, a new strong baseline model is composed, an improved dissimilarity space is constructed, a new transfer learning optimization method is proposed, and a time-space-appearance constraint is proposed. The test results show that the R1 accuracy of our model is improved by 21.4% compared with DMMD in the experiment from Market1501 to DukeMTMC.

Fusion in dissimilarity space for RGB-D person re-identification

Person re-identification (Re-id) is the task of recognizing people across different non-overlapping sensors of a camera network. Despite the recent advances with deep learning (DL) models for multi-modal fusion, state-of-the-art Re-id approaches fail to leverage the depth guided contextual information to dynamically select the most discriminant convolutional filters for better feature embedding and inference. Thanks to low cost modern RGB-D sensors (e.g. Microsoft Kinect and Intel RealSense Depth camera) that avail us with different modalities such as illumination invariant high-quality depth images, RGB images and skeleton information can be obtained simultaneously. State-of-the-art Re-id approaches utilize multi-modal fusion in feature space where the chances of fused noisy features to dominate the final recognition process is high. In this paper, we address this issue by exploiting the advantage of using an effective fusion technique in dissimilarity space. Given a query RGB-D image of an individual, two CNNs are separately trained with 3-channel RGB and 4-channel RGB-D images to produce two different feature embeddings required for pair-wise matching with embeddings for reference images, where dissimilarity scores w.r.t reference images from both modalities are fused together for final ranking. Additionally, lack of a proper RGB-D Re-id dataset prompts us to contribute a new RGB-D Re-id dataset named SUCVL RGBD-ID, including RGB and depth images of 58 identities from three cameras where one camera was installed in poor illumination conditions and the remaining two cameras were installed in two different indoor locations with different indoor lighting environments. Extensive experimental analysis on our dataset and two publicly available datasets show the effectiveness of our proposed method. Moreover, our proposed method is general and can be applied to a multitude of different RGB-D based applications.

Fusion in Dissimilarity Space Between RGB D and Skeleton for Person Re Identification

Person re-identification (Re-id) is one of the important tools of video surveillance systems, which aims to recognize an individual across the multiple disjoint sensors of a camera network. Despite the recent advances on RGB camera-based person re-identification methods under normal lighting conditions, Re-id researchers fail to take advantages of modern RGB-D sensor-based additional information (e.g. depth and skeleton information). When traditional RGB-based cameras fail to capture the video under poor illumination conditions, RGB-D sensor-based additional information can be advantageous to tackle these constraints. This work takes depth images and skeleton joint points as additional information along with RGB appearance cues and proposes a person re-identification method. We combine 4-channel RGB-D image features with skeleton information using score-level fusion strategy in dissimilarity space to increase re-identification accuracy. Moreover, our propose method overcomes the illumination problem because we use illumination invariant depth image and skeleton information. We carried out rigorous experiments on two publicly available RGBD-ID re-identification datasets and proved the use of combined features of 4-channel RGB-D images and skeleton information boost up the rank 1 recognition accuracy.

Dissimilarity space reinforced with manifold learning and latent space modeling for improved pattern classification

Dissimilarity representation plays a very important role in pattern recognition due to its ability to capture structural and relational information between samples. Dissimilarity space embedding is an approach in which each sample is represented as a vector based on its dissimilarity to some other samples called prototypes. However, lack of neighborhood-preserving, fixed and usually considerable prototype set for all training samples cause low classification accuracy and high computational complexity. To address these challenges, our proposed method creates dissimilarity space considering the neighbors of each data point on the manifold. For this purpose, Locally Linear Embedding (LLE) is used as an unsupervised manifold learning algorithm. The only goal of this step is to learn the global structure and the neighborhood of data on the manifold and mapping or dimension reduction is not performed. In order to create the dissimilarity space, each sample is compared only with its prototype set including its k-nearest neighbors on the manifold using the geodesic distance metric. Geodesic distance metric is used for the structure preserving and is computed using the weighted LLE neighborhood graph. Finally, Latent Space Model (LSM), is applied to reduce the dimensions of the Euclidean latent space so that the second challenge is resolved. To evaluate the resulted representation ad so called dissimilarity space, two common classifiers namely K Nearest Neighbor (KNN) and Support Vector Machine (SVM) are applied. Experiments on different datasets which included both Euclidean and non-Euclidean spaces, demonstrate that using the proposed approach, classifiers outperform the other basic dissimilarity spaces in both accuracy and runtime.

Closing the Performance Gap between Siamese Networks for Dissimilarity Image Classification and Convolutional Neural Networks.

In this paper, we examine two strategies for boosting the performance of ensembles of Siamese networks (SNNs) for image classification using two loss functions (Triplet and Binary Cross Entropy) and two methods for building the dissimilarity spaces (FULLY and DEEPER). With FULLY, the distance between a pattern and a prototype is calculated by comparing two images using the fully connected layer of the Siamese network. With DEEPER, each pattern is described using a deeper layer combined with dimensionality reduction. The basic design of the SNNs takes advantage of supervised k-means clustering for building the dissimilarity spaces that train a set of support vector machines, which are then combined by sum rule for a final decision. The robustness and versatility of this approach are demonstrated on several cross-domain image data sets, including a portrait data set, two bioimage and two animal vocalization data sets. Results show that the strategies employed in this work to increase the performance of dissimilarity image classification using SNN are closing the gap with standalone CNNs. Moreover, when our best system is combined with an ensemble of CNNs, the resulting performance is superior to an ensemble of CNNs, demonstrating that our new strategy is extracting additional information.

Comparison of thermal behaviour of different land cover types in Slovenia and Czech Republic

This paper presents the results of a study of thermal behaviour of different land cover types in the Czech Republic and Slovenia. A hand-held thermal camera, Fluke Ti55, was used for data collection. Variation in the values of the surface temperature characteristics reflects the geographically dissimilar spaces. The investigation demonstrated impact of dense green vegetation and water bodies on the balanced thermal behaviour of landscapes in both countries. Thus, they appear to be cold spots with the lowest values. The most obvious variation in surface temperature is associated with artificial areas and bare surfaces. They usually represent hot spots in the landscape. In both countries similar thermal behaviour was found in artificial land cover types as opposed to agricultural land cover types.

Experiments of Image Classification Using Dissimilarity Spaces Built with Siamese Networks.

Traditionally, classifiers are trained to predict patterns within a feature space. The image classification system presented here trains classifiers to predict patterns within a vector space by combining the dissimilarity spaces generated by a large set of Siamese Neural Networks (SNNs). A set of centroids from the patterns in the training data sets is calculated with supervised k-means clustering. The centroids are used to generate the dissimilarity space via the Siamese networks. The vector space descriptors are extracted by projecting patterns onto the similarity spaces, and SVMs classify an image by its dissimilarity vector. The versatility of the proposed approach in image classification is demonstrated by evaluating the system on different types of images across two domains: two medical data sets and two animal audio data sets with vocalizations represented as images (spectrograms). Results show that the proposed system’s performance competes competitively against the best-performing methods in the literature, obtaining state-of-the-art performance on one of the medical data sets, and does so without ad-hoc optimization of the clustering methods on the tested data sets.

On the Potential of Taxonomic Graphs to Improve Applicability and Performance for the Classification of Biomedical Patents

A core task in technology management in biomedical engineering and beyond is the classification of patents into domain-specific categories, increasingly automated by machine learning, with the fuzzy language of patents causing particular problems. Striving for higher classification performance, increasingly complex models have been developed, based not only on text but also on a wealth of distinct (meta) data and methods. However, this makes it difficult to access and integrate data and to fuse distinct predictions. Although the already established Cooperate Patent Classification (CPC) offers a plethora of information, it is rarely used in automated patent categorization. Thus, we combine taxonomic and textual information to an ensemble classification system comparing stacking and fixed combination rules as fusion methods. Various classifiers are trained on title/abstract and on both the CPC and IPC (International Patent Classification) assignments of 1230 patents covering six categories of future biomedical innovation. The taxonomies are modeled as tree graphs, parsed and transformed by Dissimilarity Space Embedding (DSE) to real-valued vectors. The classifier ensemble tops the basic performance by nearly 10 points to F1 = 78.7% when stacked with a feed-forward Artificial Neural Network (ANN). Taxonomic base classifiers perform nearly as well as the text-based learners. Moreover, an ensemble only of CPC and IPC learners reaches F1 = 71.2% as fully language independent and straightforward approach of established algorithms and readily available integrated data enabling new possibilities for technology management.

Animal Sound Classification Using Dissimilarity Spaces

The classifier system proposed in this work combines the dissimilarity spaces produced by a set of Siamese neural networks (SNNs) designed using four different backbones with different clustering techniques for training SVMs for automated animal audio classification. The system is evaluated on two animal audio datasets: one for cat and another for bird vocalizations. The proposed approach uses clustering methods to determine a set of centroids (in both a supervised and unsupervised fashion) from the spectrograms in the dataset. Such centroids are exploited to generate the dissimilarity space through the Siamese networks. In addition to feeding the SNNs with spectrograms, experiments process the spectrograms using the heterogeneous auto-similarities of characteristics. Once the similarity spaces are computed, each pattern is “projected” into the space to obtain a vector space representation; this descriptor is then coupled to a support vector machine (SVM) to classify a spectrogram by its dissimilarity vector. Results demonstrate that the proposed approach performs competitively (without ad-hoc optimization of the clustering methods) on both animal vocalization datasets. To further demonstrate the power of the proposed system, the best standalone approach is also evaluated on the challenging Dataset for Environmental Sound Classification (ESC50) dataset.

Modelling and Recognition of Protein Contact Networks by Multiple Kernel Learning and Dissimilarity Representations.

Multiple kernel learning is a paradigm which employs a properly constructed chain of kernel functions able to simultaneously analyse different data or different representations of the same data. In this paper, we propose an hybrid classification system based on a linear combination of multiple kernels defined over multiple dissimilarity spaces. The core of the training procedure is the joint optimisation of kernel weights and representatives selection in the dissimilarity spaces. This equips the system with a two-fold knowledge discovery phase: by analysing the weights, it is possible to check which representations are more suitable for solving the classification problem, whereas the pivotal patterns selected as representatives can give further insights on the modelled system, possibly with the help of field-experts. The proposed classification system is tested on real proteomic data in order to predict proteins’ functional role starting from their folded structure: specifically, a set of eight representations are drawn from the graph-based protein folded description. The proposed multiple kernel-based system has also been benchmarked against a clustering-based classification system also able to exploit multiple dissimilarities simultaneously. Computational results show remarkable classification capabilities and the knowledge discovery analysis is in line with current biological knowledge, suggesting the reliability of the proposed system.

Spectrogram Classification Using Dissimilarity Space

In this work, we combine a Siamese neural network and different clustering techniques to generate a dissimilarity space that is then used to train an SVM for automated animal audio classification. The animal audio datasets used are (i) birds and (ii) cat sounds, which are freely available. We exploit different clustering methods to reduce the spectrograms in the dataset to a number of centroids that are used to generate the dissimilarity space through the Siamese network. Once computed, we use the dissimilarity space to generate a vector space representation of each pattern, which is then fed into an support vector machine (SVM) to classify a spectrogram by its dissimilarity vector. Our study shows that the proposed approach based on dissimilarity space performs well on both classification problems without ad-hoc optimization of the clustering methods. Moreover, results show that the fusion of CNN-based approaches applied to the animal audio classification problem works better than the stand-alone CNNs.

A white-box analysis on the writer-independent dichotomy transformation applied to offline handwritten signature verification

High number of writers, small number of training samples per writer with high intra-class variability and heavily imbalanced class distributions are among the challenges and difficulties of the offline Handwritten Signature Verification (HSV) problem. A good alternative to tackle these issues is to use a writer-independent (WI) framework. In WI systems, a single model is trained to perform signature verification for all writers from a dissimilarity space generated by the dichotomy transformation. Among the advantages of this framework is its scalability to deal with some of these challenges and its ease in managing new writers, and hence of being used in a transfer learning context. In this work, we present a white-box analysis of this approach highlighting how it handles the challenges, the dynamic selection of references through fusion function, and its application for transfer learning. All the analyses are carried out at the instance level using the instance hardness (IH) measure. The experimental results show that, using the IH analysis, we were able to characterize “good” and “bad” quality skilled forgeries as well as the frontier region between positive and negative samples. This enables futures investigations on methods for improving discrimination between genuine signatures and skilled forgeries by considering these characterizations.

Face space representations of movement

The challenging computational problem of perceiving dynamic faces “in the wild” goes unresolved because most research focuses on easier questions about static photograph perception. This literature conceptualizes face representation as a dissimilarity-based “face space”, with axes that describe the dimensions of static images. Some versions express positions in face space relative to a central tendency (norm). Are facial movements represented like this? We tested for representations that accord with an a priori hypothesized motion-based face space by experimentally manipulating faces’ motion-based dissimilarity. Because we caricatured movements, we could test for representations of dissimilarity from a motion-based norm. Behaviorally, participants perceived these caricatured expressions as convincing and recognizable. Moreover, as expected, caricature enhanced perceived dissimilarity between facial expressions. Functional magnetic resonance imaging showed that occipitotemporal brain responses, including face-selective and motion-sensitive areas, reflect this face space. This evidence converged across methods including analysis of univariate mean responses (which additionally exhibited norm-based responses), repetition suppression and representational similarity analysis. This accumulated evidence for “representational geometry” shows how perception and visual brain responses to facial dynamics reflect representations of movement-based dissimilarity spaces, including explicit computation of distance from a norm movement.

Examining the variability in network populations and its role in generative models

AbstractA principled approach to understand networks is to formulate generative models and infer their parameters from given network data. Due to the scarcity of data in the form of multiple networks that have evolved from the same process, generative models are typically formulated to learn parameters from a single network observation, hence ignoring the natural variability of the “true” process. In this paper, we highlight the importance of variability in evaluating generative models and present two ways of quantifying the variability for a finite set of networks. The first evaluation scheme compares the statistical properties of networks in a dissimilarity space, while the other relies on data-driven entropy measures to compute variability in network populations. Using these measures, we evaluate the ability of four generative models to synthesize networks that capture the variability of the “true” process. Our empirical analysis suggests that generative models fitted for a single network observation fail to capture the variability in the network population. Our work highlights the need for rethinking the way we evaluate the goodness-of-fit of new and existing network models and devising models that are capable of matching the variability of network populations when available.

Comparison of distance-based and model-based ordinations.

Distance-based ordinations have played a critical role in community ecology for more than half a century, but are still under active development. These methods employ a matrix of pairwise distances or dissimilarities between sample units, and map sample units from the high-dimensional distance or dissimilarity space to a low-dimensional representation for analysis. Distance- or dissimilarity-based methods employ continuum or gradient ecological theory and a variety of statistical models to achieve the mapping. Recently, ecologists have developed model-based ordinations based on latent vectors and individual species response models. These methods employ the individualistic perspective of Gleason as the ecological model, and Bayesian or maximum-likelihood methods to estimate the parameters for the low dimensional space represented by the latent vectors. In this research I compared two distance-based methods (NMDS and t-SNE) with two model-based methods (BORAL and REO) on five data sets to determine which methods are superior for (1) extracting meaningful ecological drivers of variability in community composition, and (2) estimating sample unit locations in ordination space that maximize the goodness-of-fit of individual species response models to the estimated sample unit locations. Environmental variables and species were fitted to the ordinations by generalized additive models (GAMs) with Gaussian, negative binomial, or Poisson distribution models as appropriate. Across the five data sets, 22 models of environmental variability and 449 models of species distributions were calculated for each of the ordination methods. To minimize the effects of stochasticity the entire analysis was replicated three times and results averaged across the replicates. Results were evaluated by deviance explained and AIC for environmental variables and species distributions, averaged by ordination method for each data set, and ranked from best to worst. For the four assessments distance-based methods ranked 1 and 2 in three cases, and 1 and 3 in one case, significantly out performing the model-based methods. t-SNE was the top-performing method, out performing NMDS especially on the more complex data sets. In general the gradient-based theoretical basis and data sufficiency of distance-based methods allowed distance-based methods to outperform model-based methods in every assessment.