July 27, 2019

3113 words 15 mins read

Paper Group ANR 617

Face-from-Depth for Head Pose Estimation on Depth Images. General Latent Feature Modeling for Data Exploration Tasks. Causal Inference by Stochastic Complexity. Prediction of amino acid side chain conformation using a deep neural network. General Bayesian Updating and the Loss-Likelihood Bootstrap. A Minimal Solution for Two-view Focal-length Estim …

Face-from-Depth for Head Pose Estimation on Depth Images


Title	Face-from-Depth for Head Pose Estimation on Depth Images
Authors	Guido Borghi, Matteo Fabbri, Roberto Vezzani, Simone Calderara, Rita Cucchiara
Abstract	Depth cameras allow to set up reliable solutions for people monitoring and behavior understanding, especially when unstable or poor illumination conditions make unusable common RGB sensors. Therefore, we propose a complete framework for the estimation of the head and shoulder pose based on depth images only. A head detection and localization module is also included, in order to develop a complete end-to-end system. The core element of the framework is a Convolutional Neural Network, called POSEidon+, that receives as input three types of images and provides the 3D angles of the pose as output. Moreover, a Face-from-Depth component based on a Deterministic Conditional GAN model is able to hallucinate a face from the corresponding depth image. We empirically demonstrate that this positively impacts the system performances. We test the proposed framework on two public datasets, namely Biwi Kinect Head Pose and ICT-3DHP, and on Pandora, a new challenging dataset mainly inspired by the automotive setup. Experimental results show that our method overcomes several recent state-of-art works based on both intensity and depth input data, running in real-time at more than 30 frames per second.
Tasks	Head Detection, Head Pose Estimation, Pose Estimation
Published	2017-12-12
URL	http://arxiv.org/abs/1712.05277v2
PDF	http://arxiv.org/pdf/1712.05277v2.pdf
PWC	https://paperswithcode.com/paper/face-from-depth-for-head-pose-estimation-on
Repo
Framework

General Latent Feature Modeling for Data Exploration Tasks


Title	General Latent Feature Modeling for Data Exploration Tasks
Authors	Isabel Valera, Melanie F. Pradier, Zoubin Ghahramani
Abstract	This paper introduces a general Bayesian non- parametric latent feature model suitable to per- form automatic exploratory analysis of heterogeneous datasets, where the attributes describing each object can be either discrete, continuous or mixed variables. The proposed model presents several important properties. First, it accounts for heterogeneous data while can be inferred in linear time with respect to the number of objects and attributes. Second, its Bayesian nonparametric nature allows us to automatically infer the model complexity from the data, i.e., the number of features necessary to capture the latent structure in the data. Third, the latent features in the model are binary-valued variables, easing the interpretability of the obtained latent features in data exploration tasks.
Tasks
Published	2017-07-26
URL	http://arxiv.org/abs/1707.08352v1
PDF	http://arxiv.org/pdf/1707.08352v1.pdf
PWC	https://paperswithcode.com/paper/general-latent-feature-modeling-for-data
Repo
Framework

Causal Inference by Stochastic Complexity


Title	Causal Inference by Stochastic Complexity
Authors	Kailash Budhathoki, Jilles Vreeken
Abstract	The algorithmic Markov condition states that the most likely causal direction between two random variables X and Y can be identified as that direction with the lowest Kolmogorov complexity. Due to the halting problem, however, this notion is not computable. We hence propose to do causal inference by stochastic complexity. That is, we propose to approximate Kolmogorov complexity via the Minimum Description Length (MDL) principle, using a score that is mini-max optimal with regard to the model class under consideration. This means that even in an adversarial setting, such as when the true distribution is not in this class, we still obtain the optimal encoding for the data relative to the class. We instantiate this framework, which we call CISC, for pairs of univariate discrete variables, using the class of multinomial distributions. Experiments show that CISC is highly accurate on synthetic, benchmark, as well as real-world data, outperforming the state of the art by a margin, and scales extremely well with regard to sample and domain sizes.
Tasks	Causal Inference
Published	2017-02-22
URL	http://arxiv.org/abs/1702.06776v1
PDF	http://arxiv.org/pdf/1702.06776v1.pdf
PWC	https://paperswithcode.com/paper/causal-inference-by-stochastic-complexity
Repo
Framework

Prediction of amino acid side chain conformation using a deep neural network


Title	Prediction of amino acid side chain conformation using a deep neural network
Authors	Ke Liu, Xiangyan Sun, Jun Ma, Zhenyu Zhou, Qilin Dong, Shengwen Peng, Junqiu Wu, Suocheng Tan, Günter Blobel, Jie Fan
Abstract	A deep neural network based architecture was constructed to predict amino acid side chain conformation with unprecedented accuracy. Amino acid side chain conformation prediction is essential for protein homology modeling and protein design. Current widely-adopted methods use physics-based energy functions to evaluate side chain conformation. Here, using a deep neural network architecture without physics-based assumptions, we have demonstrated that side chain conformation prediction accuracy can be improved by more than 25%, especially for aromatic residues compared with current standard methods. More strikingly, the prediction method presented here is robust enough to identify individual conformational outliers from high resolution structures in a protein data bank without providing its structural factors. We envisage that our amino acid side chain predictor could be used as a quality check step for future protein structure model validation and many other potential applications such as side chain assignment in Cryo-electron microscopy, crystallography model auto-building, protein folding and small molecule ligand docking.
Tasks
Published	2017-07-26
URL	http://arxiv.org/abs/1707.08381v1
PDF	http://arxiv.org/pdf/1707.08381v1.pdf
PWC	https://paperswithcode.com/paper/prediction-of-amino-acid-side-chain
Repo
Framework

General Bayesian Updating and the Loss-Likelihood Bootstrap


Title	General Bayesian Updating and the Loss-Likelihood Bootstrap
Authors	Simon Lyddon, Chris Holmes, Stephen Walker
Abstract	In this paper we revisit the weighted likelihood bootstrap, a method that generates samples from an approximate Bayesian posterior of a parametric model. We show that the same method can be derived, without approximation, under a Bayesian nonparametric model with the parameter of interest defined as minimising an expected negative log-likelihood under an unknown sampling distribution. This interpretation enables us to extend the weighted likelihood bootstrap to posterior sampling for parameters minimizing an expected loss. We call this method the loss-likelihood bootstrap. We make a connection between this and general Bayesian updating, which is a way of updating prior belief distributions without needing to construct a global probability model, yet requires the calibration of two forms of loss function. The loss-likelihood bootstrap is used to calibrate the general Bayesian posterior by matching asymptotic Fisher information. We demonstrate the methodology on a number of examples.
Tasks	Calibration
Published	2017-09-22
URL	http://arxiv.org/abs/1709.07616v2
PDF	http://arxiv.org/pdf/1709.07616v2.pdf
PWC	https://paperswithcode.com/paper/general-bayesian-updating-and-the-loss
Repo
Framework

A Minimal Solution for Two-view Focal-length Estimation using Two Affine Correspondences


Title	A Minimal Solution for Two-view Focal-length Estimation using Two Affine Correspondences
Authors	Daniel Barath, Tekla Toth, Levente Hajder
Abstract	A minimal solution using two affine correspondences is presented to estimate the common focal length and the fundamental matrix between two semi-calibrated cameras - known intrinsic parameters except a common focal length. To the best of our knowledge, this problem is unsolved. The proposed approach extends point correspondence-based techniques with linear constraints derived from local affine transformations. The obtained multivariate polynomial system is efficiently solved by the hidden-variable technique. Observing the geometry of local affinities, we introduce novel conditions eliminating invalid roots. To select the best one out of the remaining candidates, a root selection technique is proposed outperforming the recent ones especially in case of high-level noise. The proposed 2-point algorithm is validated on both synthetic data and 104 publicly available real image pairs. A Matlab implementation of the proposed solution is included in the paper.
Tasks
Published	2017-06-06
URL	http://arxiv.org/abs/1706.01649v1
PDF	http://arxiv.org/pdf/1706.01649v1.pdf
PWC	https://paperswithcode.com/paper/a-minimal-solution-for-two-view-focal-length
Repo
Framework

Considerations of automated machine learning in clinical metabolic profiling: Altered homocysteine plasma concentration associated with metformin exposure


Title	Considerations of automated machine learning in clinical metabolic profiling: Altered homocysteine plasma concentration associated with metformin exposure
Authors	Alena Orlenko, Jason H. Moore, Patryk Orzechowski, Randal S. Olson, Junmei Cairns, Pedro J. Caraballo, Richard M. Weinshilboum, Liewei Wang, Matthew K. Breitenstein
Abstract	With the maturation of metabolomics science and proliferation of biobanks, clinical metabolic profiling is an increasingly opportunistic frontier for advancing translational clinical research. Automated Machine Learning (AutoML) approaches provide exciting opportunity to guide feature selection in agnostic metabolic profiling endeavors, where potentially thousands of independent data points must be evaluated. In previous research, AutoML using high-dimensional data of varying types has been demonstrably robust, outperforming traditional approaches. However, considerations for application in clinical metabolic profiling remain to be evaluated. Particularly, regarding the robustness of AutoML to identify and adjust for common clinical confounders. In this study, we present a focused case study regarding AutoML considerations for using the Tree-Based Optimization Tool (TPOT) in metabolic profiling of exposure to metformin in a biobank cohort. First, we propose a tandem rank-accuracy measure to guide agnostic feature selection and corresponding threshold determination in clinical metabolic profiling endeavors. Second, while AutoML, using default parameters, demonstrated potential to lack sensitivity to low-effect confounding clinical covariates, we demonstrated residual training and adjustment of metabolite features as an easily applicable approach to ensure AutoML adjustment for potential confounding characteristics. Finally, we present increased homocysteine with long-term exposure to metformin as a potentially novel, non-replicated metabolite association suggested by TPOT; an association not identified in parallel clinical metabolic profiling endeavors. While considerations are recommended, including adjustment approaches for clinical confounders, AutoML presents an exciting tool to enhance clinical metabolic profiling and advance translational research endeavors.
Tasks	AutoML, Feature Selection
Published	2017-10-09
URL	http://arxiv.org/abs/1710.03268v1
PDF	http://arxiv.org/pdf/1710.03268v1.pdf
PWC	https://paperswithcode.com/paper/considerations-of-automated-machine-learning
Repo
Framework

Video Acceleration Magnification


Title	Video Acceleration Magnification
Authors	Yichao Zhang, Silvia L. Pintea, Jan C. van Gemert
Abstract	The ability to amplify or reduce subtle image changes over time is useful in contexts such as video editing, medical video analysis, product quality control and sports. In these contexts there is often large motion present which severely distorts current video amplification methods that magnify change linearly. In this work we propose a method to cope with large motions while still magnifying small changes. We make the following two observations: i) large motions are linear on the temporal scale of the small changes; ii) small changes deviate from this linearity. We ignore linear motion and propose to magnify acceleration. Our method is pure Eulerian and does not require any optical flow, temporal alignment or region annotations. We link temporal second-order derivative filtering to spatial acceleration magnification. We apply our method to moving objects where we show motion magnification and color magnification. We provide quantitative as well as qualitative evidence for our method while comparing to the state-of-the-art.
Tasks	Optical Flow Estimation
Published	2017-04-13
URL	http://arxiv.org/abs/1704.04186v2
PDF	http://arxiv.org/pdf/1704.04186v2.pdf
PWC	https://paperswithcode.com/paper/video-acceleration-magnification
Repo
Framework

A graphical, scalable and intuitive method for the placement and the connection of biological cells


Title	A graphical, scalable and intuitive method for the placement and the connection of biological cells
Authors	Nicolas P. Rougier
Abstract	We introduce a graphical method originating from the computer graphics domain that is used for the arbitrary and intuitive placement of cells over a two-dimensional manifold. Using a bitmap image as input, where the color indicates the identity of the different structures and the alpha channel indicates the local cell density, this method guarantees a discrete distribution of cell position respecting the local density function. This method scales to any number of cells, allows to specify several different structures at once with arbitrary shapes and provides a scalable and versatile alternative to the more classical assumption of a uniform non-spatial distribution. Furthermore, several connection schemes can be derived from the paired distances between cells using either an automatic mapping or a user-defined local reference frame, providing new computational properties for the underlying model. The method is illustrated on a discrete homogeneous neural field, on the distribution of cones and rods in the retina and on a coronal view of the basal ganglia.
Tasks
Published	2017-10-14
URL	http://arxiv.org/abs/1710.05189v1
PDF	http://arxiv.org/pdf/1710.05189v1.pdf
PWC	https://paperswithcode.com/paper/a-graphical-scalable-and-intuitive-method-for
Repo
Framework

A comprehensive study of batch construction strategies for recurrent neural networks in MXNet


Title	A comprehensive study of batch construction strategies for recurrent neural networks in MXNet
Authors	Patrick Doetsch, Pavel Golik, Hermann Ney
Abstract	In this work we compare different batch construction methods for mini-batch training of recurrent neural networks. While popular implementations like TensorFlow and MXNet suggest a bucketing approach to improve the parallelization capabilities of the recurrent training process, we propose a simple ordering strategy that arranges the training sequences in a stochastic alternatingly sorted way. We compare our method to sequence bucketing as well as various other batch construction strategies on the CHiME-4 noisy speech recognition corpus. The experiments show that our alternated sorting approach is able to compete both in training time and recognition performance while being conceptually simpler to implement.
Tasks	Noisy Speech Recognition, Speech Recognition
Published	2017-05-05
URL	http://arxiv.org/abs/1705.02414v1
PDF	http://arxiv.org/pdf/1705.02414v1.pdf
PWC	https://paperswithcode.com/paper/a-comprehensive-study-of-batch-construction
Repo
Framework

Maximum entropy based non-negative optoacoustic tomographic image reconstruction


Title	Maximum entropy based non-negative optoacoustic tomographic image reconstruction
Authors	Jaya Prakash, Subhamoy Mandal, Daniel Razansky, Vasilis Ntziachristos
Abstract	Objective:Optoacoustic (photoacoustic) tomography is aimed at reconstructing maps of the initial pressure rise induced by the absorption of light pulses in tissue. In practice, due to inaccurate assumptions in the forward model, noise and other experimental factors, the images are often afflicted by artifacts, occasionally manifested as negative values. The aim of the work is to develop an inversion method which reduces the occurrence of negative values and improves the quantitative performance of optoacoustic imaging. Methods: We present a novel method for optoacoustic tomography based on an entropy maximization algorithm, which uses logarithmic regularization for attaining non-negative reconstructions. The reconstruction image quality is further improved using structural prior based fluence correction. Results: We report the performance achieved by the entropy maximization scheme on numerical simulation, experimental phantoms and in-vivo samples. Conclusion: The proposed algorithm demonstrates superior reconstruction performance by delivering non-negative pixel values with no visible distortion of anatomical structures. Significance: Our method can enable quantitative optoacoustic imaging, and has the potential to improve pre-clinical and translational imaging applications.
Tasks	Image Reconstruction
Published	2017-07-26
URL	http://arxiv.org/abs/1707.08391v3
PDF	http://arxiv.org/pdf/1707.08391v3.pdf
PWC	https://paperswithcode.com/paper/maximum-entropy-based-non-negative
Repo
Framework

Kaggle Competition: Expedia Hotel Recommendations


Title	Kaggle Competition: Expedia Hotel Recommendations
Authors	Gourav G. Shenoy, Mangirish A. Wagle, Anwar Shaikh
Abstract	With hundreds, even thousands, of hotels to choose from at every destination, it’s difficult to know which will suit your personal preferences. Expedia wants to take the proverbial rabbit hole out of hotel search by providing personalized hotel recommendations to their users. This is no small task for a site with hundreds of millions of visitors every month! Currently, Expedia uses search parameters to adjust their hotel recommendations, but there aren’t enough customer specific data to personalize them for each user. In this project, we have taken up the challenge to contextualize customer data and predict the likelihood a user will stay at 100 different hotel groups.
Tasks
Published	2017-03-06
URL	https://arxiv.org/abs/1703.02915v1
PDF	https://arxiv.org/pdf/1703.02915v1.pdf
PWC	https://paperswithcode.com/paper/kaggle-competition-expedia-hotel
Repo
Framework

Tree Projections and Constraint Optimization Problems: Fixed-Parameter Tractability and Parallel Algorithms


Title	Tree Projections and Constraint Optimization Problems: Fixed-Parameter Tractability and Parallel Algorithms
Authors	Georg Gottlob, Gianlugi Greco, Francesco Scarcello
Abstract	Tree projections provide a unifying framework to deal with most structural decomposition methods of constraint satisfaction problems (CSPs). Within this framework, a CSP instance is decomposed into a number of sub-problems, called views, whose solutions are either already available or can be computed efficiently. The goal is to arrange portions of these views in a tree-like structure, called tree projection, which determines an efficiently solvable CSP instance equivalent to the original one. Deciding whether a tree projection exists is NP-hard. Solution methods have therefore been proposed in the literature that do not require a tree projection to be given, and that either correctly decide whether the given CSP instance is satisfiable, or return that a tree projection actually does not exist. These approaches had not been generalized so far on CSP extensions for optimization problems, where the goal is to compute a solution of maximum value/minimum cost. The paper fills the gap, by exhibiting a fixed-parameter polynomial-time algorithm that either disproves the existence of tree projections or computes an optimal solution, with the parameter being the size of the expression of the objective function to be optimized over all possible solutions (and not the size of the whole constraint formula, used in related works). Tractability results are also established for the problem of returning the best K solutions. Finally, parallel algorithms for such optimization problems are proposed and analyzed. Given that the classes of acyclic hypergraphs, hypergraphs of bounded treewidth, and hypergraphs of bounded generalized hypertree width are all covered as special cases of the tree projection framework, the results in this paper directly apply to these classes. These classes are extensively considered in the CSP setting, as well as in conjunctive database query evaluation and optimization.
Tasks
Published	2017-11-14
URL	http://arxiv.org/abs/1711.05216v1
PDF	http://arxiv.org/pdf/1711.05216v1.pdf
PWC	https://paperswithcode.com/paper/tree-projections-and-constraint-optimization
Repo
Framework

Accelerating SGD for Distributed Deep-Learning Using Approximated Hessian Matrix


Title	Accelerating SGD for Distributed Deep-Learning Using Approximated Hessian Matrix
Authors	Sébastien M. R. Arnold, Chunming Wang
Abstract	We introduce a novel method to compute a rank $m$ approximation of the inverse of the Hessian matrix in the distributed regime. By leveraging the differences in gradients and parameters of multiple Workers, we are able to efficiently implement a distributed approximation of the Newton-Raphson method. We also present preliminary results which underline advantages and challenges of second-order methods for large stochastic optimization problems. In particular, our work suggests that novel strategies for combining gradients provide further information on the loss surface.
Tasks	Stochastic Optimization
Published	2017-09-15
URL	http://arxiv.org/abs/1709.05069v1
PDF	http://arxiv.org/pdf/1709.05069v1.pdf
PWC	https://paperswithcode.com/paper/accelerating-sgd-for-distributed-deep
Repo
Framework

Video Synopsis Generation Using Spatio-Temporal Groups


Title	Video Synopsis Generation Using Spatio-Temporal Groups
Authors	A. Ahmed, D. P. Dogra, S. Kar, R. Patnaik, S. Lee, H. Choi, I. Kim
Abstract	Millions of surveillance cameras operate at 24x7 generating huge amount of visual data for processing. However, retrieval of important activities from such a large data can be time consuming. Thus, researchers are working on finding solutions to present hours of visual data in a compressed, but meaningful way. Video synopsis is one of the ways to represent activities using relatively shorter duration clips. So far, two main approaches have been used by researchers to address this problem, namely synopsis by tracking moving objects and synopsis by clustering moving objects. Synopses outputs, mainly depend on tracking, segmenting, and shifting of moving objects temporally as well as spatially. In many situations, tracking fails, thus produces multiple trajectories of the same object. Due to this, the object may appear and disappear multiple times within the same synopsis output, which is misleading. This also leads to discontinuity and often can be confusing to the viewer of the synopsis. In this paper, we present a new approach for generating compressed video synopsis by grouping tracklets of moving objects. Grouping helps to generate a synopsis where chronologically related objects appear together with meaningful spatio-temporal relation. Our proposed method produces continuous, but a less confusing synopses when tested on publicly available dataset videos as well as in-house dataset videos.
Tasks
Published	2017-09-15
URL	http://arxiv.org/abs/1709.05311v1
PDF	http://arxiv.org/pdf/1709.05311v1.pdf
PWC	https://paperswithcode.com/paper/video-synopsis-generation-using-spatio
Repo
Framework