arXiv subjects

Jun Yu

Publications and source records attributed to Jun Yu.

At least 289 records · Page 16Linked to original sources

arXiv2018

Textually Guided Ranking Network for Attentional Image Retweet Modeling

Retweet prediction is a challenging problem in social media sites (SMS). In this paper, we study the problem of image retweet prediction in social media, which predicts the image sharing behavior that the user reposts the image tweets from their followees. Unlike previous studies, we learn user preference ranking model from their past retweeted image tweets in SMS. We first propose heterogeneous image retweet modeling network (IRM) that exploits users' past retweeted image tweets with associated contexts, their following relations in SMS and preference of their followees. We then develop a novel attentional multi-faceted ranking network learning framework with textually guided multi-modal neural networks for the proposed heterogenous IRM network to learn the joint image tweet representations and user preference representations for prediction task. The extensive experiments on a large-scale dataset from Twitter site shows that our method achieves better performance than other state-of-the-art solutions to the problem.

cs.IR↗

arXiv2018

Twisted root system of a (*)-subgroup

We classify (*)-subgroups of compact Lie groups of adjoint type, and associate a twisted root system to every (*)-subgroup. We study the structure of twisted root system in several aspects: properties of the small Weyl group W_{small} and its normal subgroups W_{tiny} and W_{f}; properties of finite root datum; structure of strips of infinite roots.

math.GR↗

arXiv2018

A geometric interpretation of Kirillov's Conjecture

In this article, we give a natural geometric interpretation of Kirillov's conjecture for tempered representations in the framework of orbit method. This interpretation can also be considered as a first generalization of Duflo's conjecture to non-discrete series representations.

math.RT↗

arXiv2018

Semi-supervised Hashing for Semi-Paired Cross-View Retrieval

Recently, hashing techniques have gained importance in large-scale retrieval tasks because of their retrieval speed. Most of the existing cross-view frameworks assume that data are well paired. However, the fully-paired multiview situation is not universal in real applications. The aim of the method proposed in this paper is to learn the hashing function for semi-paired cross-view retrieval tasks. To utilize the label information of partial data, we propose a semi-supervised hashing learning framework which jointly performs feature extraction and classifier learning. The experimental results on two datasets show that our method outperforms several state-of-the-art methods in terms of retrieval accuracy.

cs.CV↗

arXiv2018

On $q$-ratio CMSV for sparse recovery

Sparse recovery aims to reconstruct an unknown spare or approximately sparse signal from significantly few noisy incoherent linear measurements. As a kind of computable incoherence measure of the measurement matrix, $q$-ratio constrained minimal singular values (CMSV) was proposed in Zhou and Yu \cite{zhou2018sparse} to derive the performance bounds for sparse recovery. In this paper, we study the geometrical property of the $q$-ratio CMSV, based on which we establish new sufficient conditions for signal recovery involving both sparsity defect and measurement error. The $\ell_1$-truncated set $q$-width of the measurement matrix is developed as the geometrical characterization of $q$-ratio CMSV. In addition, we show that the $q$-ratio CMSVs of a class of structured random matrices are bounded away from zero with high probability as long as the number of measurements is large enough, therefore satisfy those established sufficient conditions. Overall, our results generalize the results in Zhang and Cheng \cite{zc} from $q=2$ to any $q\in(1,\infty]$ and complement the arguments of $q$-ratio CMSV from a geometrical view.

cs.IT↗

arXiv2018

Rethinking Diversified and Discriminative Proposal Generation for Visual Grounding

Visual grounding aims to localize an object in an image referred to by a textual query phrase. Various visual grounding approaches have been proposed, and the problem can be modularized into a general framework: proposal generation, multi-modal feature representation, and proposal ranking. Of these three modules, most existing approaches focus on the latter two, with the importance of proposal generation generally neglected. In this paper, we rethink the problem of what properties make a good proposal generator. We introduce the diversity and discrimination simultaneously when generating proposals, and in doing so propose Diversified and Discriminative Proposal Networks model (DDPN). Based on the proposals generated by DDPN, we propose a high performance baseline model for visual grounding and evaluate it on four benchmark datasets. Experimental results demonstrate that our model delivers significant improvements on all the tested data-sets (e.g., 18.8\% improvement on ReferItGame and 8.2\% improvement on Flickr30k Entities over the existing state-of-the-arts respectively)

cs.CV↗

arXiv2018

FishEyeRecNet: A Multi-Context Collaborative Deep Network for Fisheye Image Rectification

Images captured by fisheye lenses violate the pinhole camera assumption and suffer from distortions. Rectification of fisheye images is therefore a crucial preprocessing step for many computer vision applications. In this paper, we propose an end-to-end multi-context collaborative deep network for removing distortions from single fisheye images. In contrast to conventional approaches, which focus on extracting hand-crafted features from input images, our method learns high-level semantics and low-level appearance features simultaneously to estimate the distortion parameters. To facilitate training, we construct a synthesized dataset that covers various scenes and distortion parameter settings. Experiments on both synthesized and real-world datasets show that the proposed model significantly outperforms current state of the art methods. Our code and synthesized dataset will be made publicly available.

cs.CV↗

arXiv2018

On the dimension datum of a subgroup. II

This paper studies three aspects around dimension datum: (1), a generalization of the dimension datum, which we call the tau-dimension datum; (2), dimension data of disconnected subgroups; (3), compactness of isospectral sets of normal homogeneous spaces.

math.GR↗

arXiv2018

Sparse recovery based on q-ratio constrained minimal singular values

We study verifiable sufficient conditions and computable performance bounds for sparse recovery algorithms such as the Basis Pursuit, the Dantzig selector and the Lasso estimator, in terms of a newly defined family of quality measures for the measurement matrices. With high probability, the developed measures for subgaussian random matrices are bounded away from zero as long as the number of measurements is reasonably large. Comparing to the restricted isotropic constant based performance analysis, the arguments in this paper are much more concise and the obtained bounds are tighter. Numerical experiments are presented to illustrate our theoretical results.

cs.IT↗

arXiv2018

A New Wald Test for Hypothesis Testing Based on MCMC outputs

In this paper, a new and convenient $χ^2$ wald test based on MCMC outputs is proposed for hypothesis testing. The new statistic can be explained as MCMC version of Wald test and has several important advantages that make it very convenient in practical applications. First, it is well-defined under improper prior distributions and avoids Jeffrey-Lindley's paradox. Second, it's asymptotic distribution can be proved to follow the $χ^2$ distribution so that the threshold values can be easily calibrated from this distribution. Third, it's statistical error can be derived using the Markov chain Monte Carlo (MCMC) approach. Fourth, most importantly, it is only based on the posterior MCMC random samples drawn from the posterior distribution. Hence, it is only the by-product of the posterior outputs and very easy to compute. In addition, when the prior information is available, the finite sample theory is derived for the proposed test statistic. At last, the usefulness of the test is illustrated with several applications to latent variable models widely used in economics and finance.

econ.EM↗

arXiv2017

Multi-modal Face Pose Estimation with Multi-task Manifold Deep Learning

Human face pose estimation aims at estimating the gazing direction or head postures with 2D images. It gives some very important information such as communicative gestures, saliency detection and so on, which attracts plenty of attention recently. However, it is challenging because of complex background, various orientations and face appearance visibility. Therefore, a descriptive representation of face images and mapping it to poses are critical. In this paper, we make use of multi-modal data and propose a novel face pose estimation method that uses a novel deep learning framework named Multi-task Manifold Deep Learning $M^2DL$. It is based on feature extraction with improved deep neural networks and multi-modal mapping relationship with multi-task learning. In the proposed deep learning based framework, Manifold Regularized Convolutional Layers (MRCL) improve traditional convolutional layers by learning the relationship among outputs of neurons. Besides, in the proposed mapping relationship learning method, different modals of face representations are naturally combined to learn the mapping function from face images to poses. In this way, the computed mapping model with multiple tasks is improved. Experimental results on three challenging benchmark datasets DPOSE, HPID and BKHPD demonstrate the outstanding performance of $M^2DL$.

cs.CV↗

arXiv2017

Phaseless compressive sensing using partial support information

We study the recovery conditions of weighted $\ell_1$ minimization for real-valued signal reconstruction from phaseless compressive sensing measurements when partial support information is available. A strong restricted isometry property condition is provided to ensure the stable recovery. Moreover, we present the weighted null space property as the sufficient and necessary condition for the success of $k$-sparse phaseless recovery via weighted $\ell_1$ minimization. Numerical experiments are conducted to illustrate our results.

cs.IT↗

arXiv2017

Recovery analysis for weighted mixed $\ell_2/\ell_p$ minimization with $0<p\leq 1$

We study the recovery conditions of weighted mixed $\ell_2/\ell_p\,(0<p\leq 1)$ minimization for block sparse signal reconstruction from compressed measurements when partial block support information is available. We show that the block $p$-restricted isometry property (RIP) can ensure the robust recovery. Moreover, we present the sufficient and necessary condition for the recovery by using weighted block $p$-null space property. The relationship between the block $p$-RIP and the weighted block $p$-null space property has been established. Finally, we illustrate our results with a series of numerical experiments.

cs.IT↗

arXiv2017

Sparsity estimation in compressive sensing with application to MR images

The theory of compressive sensing (CS) asserts that an unknown signal $\mathbf{x} \in \mathbb{C}^N$ can be accurately recovered from $m$ measurements with $m\ll N$ provided that $\mathbf{x}$ is sparse. Most of the recovery algorithms need the sparsity $s=\lVert\mathbf{x}\rVert_0$ as an input. However, generally $s$ is unknown, and directly estimating the sparsity has been an open problem. In this study, an estimator of sparsity is proposed by using Bayesian hierarchical model. Its statistical properties such as unbiasedness and asymptotic normality are proved. In the simulation study and real data study, magnetic resonance image data is used as input signal, which becomes sparse after sparsified transformation. The results from the simulation study confirm the theoretical properties of the estimator. In practice, the estimate from a real MR image can be used for recovering future MR images under the framework of CS if they are believed to have the same sparsity level after sparsification.

stat.ME↗

arXiv2017

Stable and robust $\ell_p$-constrained compressive sensing recovery via robust width property

We study the recovery results of $\ell_p$-constrained compressive sensing (CS) with $p\geq 1$ via robust width property and determine conditions on the number of measurements for standard Gaussian matrices under which the property holds with high probability. Our paper extends the existing results in Cahill and Mixon (2014) from $\ell_2$-constrained CS to $\ell_p$-constrained case with $p\geq 1$ and complements the recovery analysis for robust CS with $\ell_p$ loss function.

cs.IT↗

arXiv2017

Multi-modal Factorized Bilinear Pooling with Co-Attention Learning for Visual Question Answering

Visual question answering (VQA) is challenging because it requires a simultaneous understanding of both the visual content of images and the textual content of questions. The approaches used to represent the images and questions in a fine-grained manner and questions and to fuse these multi-modal features play key roles in performance. Bilinear pooling based models have been shown to outperform traditional linear models for VQA, but their high-dimensional representations and high computational complexity may seriously limit their applicability in practice. For multi-modal feature fusion, here we develop a Multi-modal Factorized Bilinear (MFB) pooling approach to efficiently and effectively combine multi-modal features, which results in superior performance for VQA compared with other bilinear pooling approaches. For fine-grained image and question representation, we develop a co-attention mechanism using an end-to-end deep network architecture to jointly learn both the image and question attentions. Combining the proposed MFB approach with co-attention learning in a new network architecture provides a unified model for VQA. Our experimental results demonstrate that the single MFB with co-attention model achieves new state-of-the-art performance on the real-world VQA dataset. Code available at https://github.com/yuzcccc/mfb.

cs.CV↗

arXiv2017

Embedding Visual Hierarchy with Deep Networks for Large-Scale Visual Recognition

In this paper, a level-wise mixture model (LMM) is developed by embedding visual hierarchy with deep networks to support large-scale visual recognition (i.e., recognizing thousands or even tens of thousands of object classes), and a Bayesian approach is used to adapt a pre-trained visual hierarchy automatically to the improvements of deep features (that are used for image and object class representation) when more representative deep networks are learned along the time. Our LMM model can provide an end-to-end approach for jointly learning: (a) the deep networks to extract more discriminative deep features for image and object class representation; (b) the tree classifier for recognizing large numbers of object classes hierarchically; and (c) the visual hierarchy adaptation for achieving more accurate indexing of large numbers of object classes hierarchically. By supporting joint learning of the tree classifier, the deep networks and the visual hierarchy adaptation, our LMM algorithm can provide an effective approach for controlling inter-level error propagation effectively, thus it can achieve better accuracy rates on large-scale visual recognition. Our experiments are carried on ImageNet1K and ImageNet10K image sets, and our LMM algorithm can achieve very competitive results on both the accuracy rates and the computation efficiency as compared with the baseline methods.

cs.CV↗

arXiv2017

Deep Mixture of Diverse Experts for Large-Scale Visual Recognition

In this paper, a deep mixture of diverse experts algorithm is developed for seamlessly combining a set of base deep CNNs (convolutional neural networks) with diverse outputs (task spaces), e.g., such base deep CNNs are trained to recognize different subsets of tens of thousands of atomic object classes. First, a two-layer (category layer and object class layer) ontology is constructed to achieve more effective solution for task group generation, e.g., assigning the semantically-related atomic object classes at the sibling leaf nodes into the same task group because they may share similar learning complexities. Second, one particular base deep CNNs with $M+1$ ($M \leq 1,000$) outputs is learned for each task group to recognize its $M$ atomic object classes effectively and identify one special class of "not-in-group" automatically, and the network structure (numbers of layers and units in each layer) of the well-designed AlexNet is directly used to configure such base deep CNNs. A deep multi-task learning algorithm is developed to leverage the inter-class visual similarities to learn more discriminative base deep CNNs and multi-task softmax for enhancing the separability of the atomic object classes in the same task group. Finally, all these base deep CNNs with diverse outputs (task spaces) are seamlessly combined to form a deep mixture of diverse experts for recognizing tens of thousands of atomic object classes. Our experimental results have demonstrated that our deep mixture of diverse experts algorithm can achieve very competitive results on large-scale visual recognition.

cs.CV↗