Search

One-class learning is the classic problem of fitting a model to the data for which annotations are available only for a single class. In this paper, we explore novel objectives for one-class learning, which we collectively refer to as Generalized One-class Discriminative Subspaces (GODS). Our key idea is to learn a pair of complementary classifiers to flexibly bound the one-class data distribution, where the data belongs to the positive half-space of one of the classifiers in the complementary pair and to the negative half-space of the other. To avoid redundancy while allowing non-linearity in the classifier decision surfaces, we propose to design each classifier as an orthonormal frame and seek to learn these frames via jointly optimizing for two conflicting objectives, namely: i) to minimize the distance between the two frames, and ii) to maximize the margin between the frames and the data. The learned orthonormal frames will thus characterize a piecewise linear decision surface that allows for efficient inference, while our objectives seek to bound the data within a minimal volume that maximizes the decision margin, thereby robustly capturing the data distribution. We explore several variants of our formulation under different constraints on the constituent classifiers, including kernelized feature maps. We demonstrate the empirical benefits of our approach via experiments on data from several applications in computer vision, such as anomaly detection in video sequences, human poses, and human activities. We also explore the generality and effectiveness of GODS for non-vision tasks via experiments on several UCI datasets, demonstrating state-of-the-art results. [ABSTRACT FROM AUTHOR]

Representations in the form of Symmetric Positive Definite (SPD) matrices have been popularized in a variety of visual learning applications due to their demonstrated ability to capture rich second-order statistics of visual data. There exist several similarity measures for comparing SPD matrices with documented benefits. However, selecting an appropriate measure for a given problem remains a challenge and in most cases, is the result of a trial-and-error process. In this paper, we propose to learn similarity measures in a data-driven manner. To this end, we capitalize on the $\alpha \beta$ α β -log-det divergence, which is a meta-divergence parametrized by scalars $\alpha$ α and $\beta$ β , subsuming a wide family of popular information divergences on SPD matrices for distinct and discrete values of these parameters. Our key idea is to cast these parameters in a continuum and learn them from data. We systematically extend this idea to learn vector-valued parameters, thereby increasing the expressiveness of the underlying non-linear measure. We conjoin the divergence learning problem with several standard tasks in machine learning, including supervised discriminative dictionary learning and unsupervised SPD matrix clustering. We present Riemannian gradient descent schemes for optimizing our formulations efficiently, and show the usefulness of our method on eight standard computer vision tasks. [ABSTRACT FROM AUTHOR]

DNA‐free genome editing involves the direct introduction of ribonucleoprotein (RNP) complexes into cells, but this strategy has rarely been successful in plants. In the present study, we describe a new technique for the introduction of RNPs into plant cells involving the generation of cavitation bubbles using a pulsed laser. The resulting shockwave achieves the efficient transfection of walled cells in tissue explants by creating transient membrane pores. RNP‐containing cells were rapidly identified by fluorescence microscopy, followed by regeneration and the screening of mutant plants by high‐resolution melt analysis. We used this technique in Nicotiana tabacum to target the endogenous phytoene desaturase (PDS) and actin depolymerizing factor (ADF) genes. Genome‐edited plants were produced with an efficiency of 35.2% for PDS and 16.5% for ADF. Further we evaluated the physiological, cellular and molecular effects of ADF mutations in T2 mutant plants under drought and salinity stress. The results suggest that ADF acts as a key regulator of osmotic stress tolerance in plants. [ABSTRACT FROM AUTHOR]

Most popular deep models for action recognition in videos generate independent predictions for short clips, which are then pooled heuristically to assign an action label to the full video segment. As not all frames may characterize the underlying action—indeed, many are common across multiple actions—pooling schemes that impose equal importance on all frames might be unfavorable. In an attempt to tackle this problem, we propose discriminative pooling, based on the notion that among the deep features generated on all short clips, there is at least one that characterizes the action. To identify these useful features, we resort to a negative bag consisting of features that are known to be irrelevant, for example, they are sampled either from datasets that are unrelated to our actions of interest or are CNN features produced via random noise as input. With the features from the video as a positive bag and the irrelevant features as the negative bag, we cast an objective to learn a (nonlinear) hyperplane that separates the unknown useful features from the rest in a multiple instance learning formulation within a support vector machine setup. We use the parameters of this separating hyperplane as a descriptor for the full video segment. Since these parameters are directly related to the support vectors in a max-margin framework, they can be treated as a weighted average pooling of the features from the bags, with zero weights given to non-support vectors. Our pooling scheme is end-to-end trainable within a deep learning framework. We report results from experiments on eight computer vision benchmark datasets spanning a variety of video-related tasks and demonstrate state-of-the-art performance across these tasks. [ABSTRACT FROM AUTHOR]

Deep learning models for video-based action recognition usually generate features for short clips (consisting of a few frames); such clip-level features are aggregated to video-level representations by computing statistics on these features. Typically zero-th (max) or the first-order (average) statistics are used. In this paper, we explore the benefits of using second-order statistics.Specifically, we propose a novel end-to-end learnable feature aggregation scheme, dubbed temporal correlation pooling that generates an action descriptor for a video sequence by capturing the similarities between the temporal evolution of clip-level CNN features computed across the video. Such a descriptor, while being computationally cheap, also naturally encodes the co-activations of multiple CNN features, thereby providing a richer characterization of actions than their first-order counterparts. We also propose higher-order extensions of this scheme by computing correlations after embedding the CNN features in a reproducing kernel Hilbert space. We provide experiments on benchmark datasets such as HMDB-51 and UCF-101, fine-grained datasets such as MPII Cooking activities and JHMDB, as well as the recent Kinetics-600. Our results demonstrate the advantages of higher-order pooling schemes that when combined with hand-crafted features (as is standard practice) achieves state-of-the-art accuracy. [ABSTRACT FROM AUTHOR]

Most of the existing works on fine-grained image categorization and retrieval focus on finding similar images from the same species and often give little importance to inter-species similarities. However, these similarities may carry species correlations such as the same ancestors or similar habits, which are helpful in taxonomy and understanding biological traits. In this paper, we devise a new fine-grained retrieval task that searches for similar instances from different species based on body parts. To this end, we propose a two-step strategy. In the first step, we search for visually similar parts to a query image using a deep convolutional neural network (CNN). To improve the quality of the retrieved candidates, structural cues are introduced into the CNN using a novel part-pooling layer, in which the receptive field of each part is adjusted automatically. In the second step, we re-rank the retrieved candidates to improve the species diversity. We achieve this by formulating a novel ranking function that balances between the similarity of the candidates to the queried parts, while decreasing the similarity to the query species. We provide experiments on the benchmark CUB200 dataset and Columbia Dogs dataset, and demonstrate clear benefits of our schemes. [ABSTRACT FROM AUTHOR]

During cardiac trabeculation, cardiomyocytes delaminate from the outermost (compact) layer to form complex muscular structures known as trabeculae. As these cardiomyocytes delaminate, the remodeling of adhesion junctions must be tightly coordinated so cells can extrude from the compact layer while remaining in tight contact with their neighbors. In this study, we examined the distribution of N-cadherin (Cdh2) during cardiac trabeculation in zebrafish. By analyzing the localization of a Cdh2-EGFP fusion protein expressed under the control of the zebrafish cdh2 promoter, we initially observed Cdh2-EGFP expression along the lateral sides of embryonic cardiomyocytes, in an evenly distributed pattern, and with the occasional appearance of punctae. Within a few hours, Cdh2-EGFP distribution on the lateral sides of cardiomyocytes evolves into a clear punctate pattern as Cdh2-EGFP molecules outside the punctae cluster to increase the size of these aggregates. In addition, Cdh2-EGFP molecules also appear on the basal side of cardiomyocytes that remain in the compact layer. Delaminating cardiomyocytes accumulate Cdh2-EGFP on the surface facing the basal side of compact layer cardiomyocytes, thereby allowing tight adhesion between these layers. Importantly, we find that blood flow/cardiac contractility is required for the transition from an even distribution of Cdh2-EGFP to the formation of punctae. Furthermore, using time-lapse imaging of beating hearts in conjunction with a Cdh2 tandem fluorescent protein timer transgenic line, we observed that Cdh2-EGFPmolecules appear to move from the lateral to the basal side of cardiomyocytes along the cell membrane, and that Erb-b2 receptor tyrosine kinase 2 (Erbb2) function is required for this relocalization. [ABSTRACT FROM AUTHOR]

Symmetric Positive Definite (SPD) matrices emerge as data descriptors in several applications of computer vision such as object tracking, texture recognition, and diffusion tensor imaging. Clustering these data matrices forms an integral part of these applications, for which soft-clustering algorithms (K-Means, expectation maximization, etc.) are generally used. As is well-known, these algorithms need the number of clusters to be specified, which is difficult when the dataset scales. To address this issue, we resort to the classical nonparametric Bayesian framework by modeling the data as a mixture model using the Dirichlet process (DP) prior. Since these matrices do not conform to the Euclidean geometry, rather belongs to a curved Riemannian manifold,existing DP models cannot be directly applied. Thus, in this paper, we propose a novel DP mixture model framework for SPD matrices. Using the log-determinant divergence as the underlying dissimilarity measure to compare these matrices, and further using the connection between this measure and the Wishart distribution, we derive a novel DPM model based on the Wishart-Inverse-Wishart conjugate pair. We apply this model to several applications in computer vision. Our experiments demonstrate that our model is scalable to the dataset size and at the same time achieves superior accuracy compared to several state-of-the-art parametric and nonparametric clustering algorithms. [ABSTRACT FROM PUBLISHER]

Plant growth during abiotic stress is a long sought-after trait especially in crop plants in the context of global warming and climate change. Previous studies on leaf epidermal cells have revealed that during normal growth and development, adjacent cells interdigitate anisotropically to form cell morphological patterns known as interlocking marginal lobes (IMLs), involving the cell wall-cell membrane-cortical actin continuum. IMLs are growth-associated cell morphological changes in which auxin-binding protein (ABP), Rho GTPases and actin are known to play important roles. In the present study, we investigated the formation of IMLs under drought stress and found that Erianthus arundinaceus, a drought-tolerant wild relative of sugarcane, develops such growth-related cell morphological patterns under drought stress. Using confocal microscopy, we showed an increasing trend in cortical F-actin intensity in drought-tolerant plants with increasing soil moisture stress. In order to check the role of drought tolerance-related genes in IML formation under soil moisture stress, we adopted a structural data mining strategy and identified indirect connections between the ABPs and heat shock proteins (HSPs). Initial experimental evidence for this connection comes from the high transcript levels of HSP70 observed in drought-stressed Erianthus, which developed anisotropic interdigitation, i.e. IMLs. Subsequently, by overexpressing the E. arundinaceus HSP70 gene (EaHSP70) in sugarcane (Saccharum spp. hybrid), we confirm the role of HSP70 in the formation of anisotropic interdigitation under drought stress. Taken together, our results suggest that EaHSP70 acts as a key regulator in the formation of anisotropic interdigitation in drought-tolerant plants (Erianthus and HSP70 transgenic sugarcane) under moisture stress in an actin-mediated pathway. The possible biological significance of the formation of drought-associated interlocking marginal lobes (DaIMLs) in sugarcane plants upon drought stress is discussed. [ABSTRACT FROM AUTHOR]

Although cortical actin plays an important role in cellular mechanics and morphogenesis, there is surprisingly little information on cortex organization at the apical surface of cells. In this paper, we characterize organization and dynamics of microvilli (MV) and a previously unappreciated actomyosin network at the apical surface of Madin-Darby canine kidney cells. In contrast to short and static MV in confluent cells, the apical surfaces of nonconfluent epithelial cells (ECs) form highly dynamic protrusions, which are often oriented along the plane of the membrane. These dynamic MV exhibit complex and spatially correlated reorganization, which is dependent on myosin II activity. Surprisingly, myosin II is organized into an extensive network of filaments spanning the entire apical membrane in nonconfluent ECs. Dynamic MV, myosin filaments, and their associated actin filaments form an interconnected, pre-stressed network. Interestingly, this network regulates lateral mobility of apical membrane probes such as integrins or epidermal growth factor receptors, suggesting that coordinated actomyosin dynamics contributes to apical cell membrane organization. [ABSTRACT FROM AUTHOR]

Covariance matrices have found success in several computer vision applications, including activity recognition, visual surveillance, and diffusion tensor imaging. This is because they provide an easy platform for fusing multiple features compactly. An important task in all of these applications is to compare two covariance matrices using a (dis)similarity function, for which the common choice is the Riemannian metric on the manifold inhabited by these matrices. As this Riemannian manifold is not flat, the dissimilarities should take into account the curvature of the manifold. As a result, such distance computations tend to slow down, especially when the matrix dimensions are large or gradients are required. Further, suitability of the metric to enable efficient nearest neighbor retrieval is an important requirement in the contemporary times of big data analytics. To alleviate these difficulties, this paper proposes a novel dissimilarity measure for covariances, the Jensen-Bregman LogDet Divergence (JBLD). This divergence enjoys several desirable theoretical properties and at the same time is computationally less demanding (compared to standard measures). Utilizing the fact that the square root of JBLD is a metric, we address the problem of efficient nearest neighbor retrieval on large covariance datasets via a metric tree data structure. To this end, we propose a K-Means clustering algorithm on JBLD. We demonstrate the superior performance of JBLD on covariance datasets from several computer vision applications. [ABSTRACT FROM AUTHOR]

Human actions in video sequences are characterized by the complex interplay between spatial features and their temporal dynamics. In this paper, we propose novel tensor representations for compactly capturing such higher-order relationships between visual features for the task of action recognition. We propose two tensor-based feature representations, viz. (i) sequence compatibility kernel (SCK) and (ii) dynamics compatibility kernel (DCK). SCK builds on the spatio-temporal correlations between features, whereas DCK explicitly models the action dynamics of a sequence. We also explore generalization of SCK, coined SCK $\;\oplus$ ⊕ , that operates on subsequences to capture the local-global interplay of correlations, which can incorporate multi-modal inputs e.g., skeleton 3D body-joints and per-frame classifier scores obtained from deep learning models trained on videos. We introduce linearization of these kernels that lead to compact and fast descriptors. We provide experiments on (i) 3D skeleton action sequences, (ii) fine-grained video sequences, and (iii) standard non-fine-grained videos. As our final representations are tensors that capture higher-order relationships of features, they relate to co-occurrences for robust fine-grained recognition (Lin, 2017), (Koniusz, 2018). We use higher-order tensors and so-called Eigenvalue Power Normalization (EPN) which have been long speculated to perform spectral detection of higher-order occurrences (Koniusz, 2013), (Koniusz, 2017), thus detecting fine-grained relationships of features rather than merely count features in action sequences. We prove that a tensor of order $r$ r , built from $Z_*$ Z * dimensional features, coupled with EPN indeed detects if at least one higher-order occurrence is ‘projected’ into one of its $\binom{Z_*}{r}$ Z * r subspaces of dim. $r$ r represented by the tensor, thus forming a Tensor Power Normalization metric endowed with $\binom{Z_*}{r}$ Z * r such ‘detectors’. [ABSTRACT FROM AUTHOR]

We present a principled approach to uncover the structure of visual data by solving a deep learning task coined visual permutation learning. The goal of this task is to find the permutation that recovers the structure of data from shuffled versions of it. In the case of natural images, this task boils down to recovering the original image from patches shuffled by an unknown permutation matrix. Permutation matrices are discrete, thereby posing difficulties for gradient-based optimization methods. To this end, we resort to a continuous approximation using doubly-stochastic matrices and formulate a novel bi-level optimization problem on such matrices that learns to recover the permutation. Unfortunately, such a scheme leads to expensive gradient computations. We circumvent this issue by further proposing a computationally cheap scheme for generating doubly stochastic matrices based on Sinkhorn iterations. To implement our approach we propose DeepPermNet, an end-to-end CNN model for this task. The utility of DeepPermNet is demonstrated on three challenging computer vision problems, namely, relative attributes learning, supervised learning-to-rank, and self-supervised representation learning. Our results show state-of-the-art performance on the Public Figures and OSR benchmarks for relative attributes learning, chronological and interestingness image ranking for supervised learning-to-rank, and competitive results in the classification and segmentation tasks of the PASCAL VOC dataset for self-supervised representation learning. [ABSTRACT FROM AUTHOR]

Organ-on-a-chip (OOAC) is an emerging technology based on microfluid platforms and in vitro cell culture that has a promising future in the healthcare industry. The numerous advantages of OOAC over conventional systems make it highly popular. The chip is an innovative combination of novel technologies, including lab-on-a-chip, microfluidics, biomaterials, and tissue engineering. This paper begins by analyzing the need for the development of OOAC followed by a brief introduction to the technology. Later sections discuss and review the various types of OOACs and the fabrication materials used. The implementation of artificial intelligence in the system makes it more advanced, thereby helping to provide a more accurate diagnosis as well as convenient data management. We introduce selected OOAC projects, including applications to organ/disease modelling, pharmacology, personalized medicine, and dentistry. Finally, we point out certain challenges that need to be surmounted in order to further develop and upgrade the current systems. [ABSTRACT FROM AUTHOR]

Unsupervised classification plays an important role in understanding polarimetric synthetic aperture radar (PolSAR) images. One of the typical representations of PolSAR data is in the form of Hermitian positive definite (HPD) covariance matrices. Most algorithms for unsupervised classification using this representation either use statistical distribution models or adopt polarimetric target decompositions. In this paper, we propose an unsupervised classification method by introducing a sparsity-based similarity measure on HPD matrices. Specifically, we first use a novel Riemannian sparse coding scheme for representing each HPD covariance matrix as sparse linear combinations of other HPD matrices, where the sparse reconstruction loss is defined by the Riemannian geodesic distance between HPD matrices. The coefficient vectors generated by this step reflect the neighborhood structure of HPD matrices embedded in the Euclidean space and hence can be used to define a similarity measure. We apply the scheme for PolSAR data, in which we first oversegment the images into superpixels, followed by representing each superpixel by an HPD matrix. These HPD matrices are then sparse coded, and the resulting sparse coefficient vectors are then clustered by spectral clustering using the neighborhood matrix generated by our similarity measure. The experimental results on different fully PolSAR images demonstrate the superior performance of the proposed classification approach against the state-of-the-art approaches. [ABSTRACT FROM PUBLISHER]