1,708 Works

Voice Conversion using GMM with Enhanced Global Variance

Hadas Benisty & David Malah
The goal of voice conversion is to transform a sentence said by one speaker, to sound as if another speaker had said it. The classical conversion based on a Gaussian Mixture Model and several other schemes suggested since, produce muffled sounding outputs, due to excessive smoothing of the spectral envelopes. To reduce the muffling effect, enhancement of the Global Variance (GV) of the spectral features was recently suggested. We propose a different approach for GV...

PCA saliency

Ran Margolin, & Lihi Zelnik-Manor
What makes an object salient? Most previous work assert that distinctness is the dominating factor. The difference between the various algorithms is in the way they compute distinctness. Some focus on the patterns, others on the colors, and several add high-level cues and priors. We propose a simple, yet powerful, algorithm that integrates these three factors. Our key contribution is a novel and fast approach to compute pattern distinctness. We rely on the inner statistics...

Combo Synergy of Two Drugs

Jennifer O’Neil, Yair Benita, Igor Feldman, Melissa Chenard, Brian Roberts, , Jing Li, Astrid Kral, Serguei Lejnine, Andrey Loboda, William Arthur, , Brian B. Haines, Christopher Winter, Theresa Zhang, Andrew Bloecher & Stuart D. Shumway
Combination drug therapy is a widely used paradigm for managing numerous human malignancies. In cancer treatment, additive and/or synergistic drug combinations can convert weakly efficacious monotherapies into regimens that produce robust anti-tumor activity. This can be explained in part through pathway interdependencies that are critical for cancer cell proliferation and survival. However, identification of the various interdependencies is difficult due to the complex molecular circuitry that underlies tumor development and progression. Here, we present a...

Photometric Ambient Occlusion for Intrinsic Image Decomposition

Daniel Hauagge, Scott Wehrwein, Kavita Bala & Noah Snavely
We present a method for computing ambient occlusion (AO) for a stack of images of a Lambertian scene from a fixed viewpoint. Ambient occlusion, a concept common in computer graphics, characterizes the local visibility at a point: it approximates how much light can reach that point from different directions without getting blocked by other geometry. While AO has received surprisingly little attention in vision, we show that it can be approximated using simple, per-pixel statistics...

Multiscale Anomaly Detection Using Diffusion Maps

Gal Mishne & Israel Cohen
We propose a multiscale approach to anomaly detection in images, combining spectral dimensionality reduction and a nearest-neighbor-based anomaly score. We use diffusion maps to embed the data in a low dimensional representation, which separates the anomaly from the background. The diffusion distance between points is then used to estimate the local density of each pixel in the new embedding. The diffusion map is constructed based on a subset of samples from the image and then...


Ruth Heller, Marina Bogomolov & Yoav Benjamini
The use of big data is becoming a central way of discovering knowledge in modern science. Large amounts of potential findings are screened to discover the few real ones. To verify these discoveries a follow-up study is often conducted, wherein only the promising discoveries are followed up. Such follow-up studies are common in genomics, in proteomics, and in other areas where high-throughput methods are used. We show how to decide whether promising findings from the...

A q-Quantile Estimator for High-Dimensional Distributions

Assaf Glazer, Michael Lindenbaum & Shaul Markovitch
In this paper we introduce a novel method that can efficiently estimate a family of hierarchical dense sets in high-dimensional distributions. Our method can be regarded as a natural extension of the one-class SVM (OCSVM) algorithm that finds multiple parallel separating hyperplanes in a reproducing kernel Hilbert space. We call our method q-OCSVM, as it can be used to estimate q quantiles of a highdimensional distribution. For this purpose, we introduce a new global convex...

Feature Selection Based on High Dimensional Model Representation

, Hüseyin Kaya & Lorenzo Bruzzone
This project provides the implementation of the algorithm proposed in the article "Feature Selection Based on High Dimensional Model Representation for Hyperspectral Images" (DOI:10.1109/TIP.2017.2687128). The project also contains the implementations of the other feature selection algorithms used in the article. In order to give the users the ability to recreate the results in article, the datasets and a couple of visualisation scripts are also provided. By running the algorithm, you will use eight different feature...

UQP heuristic methods

Here, several heuristic methods to solve UQP are included: domeig.m contains the dominant eigenvector matching heuristic greedy.m contains the greedy strategy rswapgreedy.m contains the row-swap greedy strategy sdprel.m contains the semidefinite relaxation method pwr.m contains the power method sim_run.m is the main file, which calls each of the above methods and compares the performance. Preprint available at http://www.optimization-online.org/DB_FILE/2016/12/5763.pdf

Reanalysis of mouse ENCODE comparative gene expression data

Yoav Gilad & Orna Mizrahi-Man
The Mouse ENCODE Consortium reported that comparative gene expression data from human and mouse tend to cluster more by species rather than by tissue. This observation was surprising, as it contradicted much of the comparative gene regulatory data collected previously, as well as the common notion that major developmental pathways are highly conserved across a wide range of species, in particular across mammals. Here we show that the Mouse ENCODE gene expression data were collected...

The FEC Algorithm for Community Mining from a Signed Network

Bo Yang, William K. Cheung &
This code is used for community mining from signed networks. The algorithm contains two main phases: 1) Finding a community (FC), 2) Extracting the Sink Community (EC).

Robust Estimation of Self-Exciting Generalized Linear Models

Abbas Kazemipour, Min Wu &
This is an implementation of the algorithms in MATLAB 2015b by Abbas Kazemipour and requires installation of the CVX package. Distribution of this code is allowed for non-commercial purposes as long as it contains the readme file.

Code for \"Coral Sr-U thermometry\"

Thomas DeCarlo
This script analyses coral trace element ratios following DeCarlo et al. (2016). Measured skeletal Sr/Ca and U/Ca ratios are combined in a new metric "Sr-U" that is strongly related to temperature for 14 IndoPacific Porites corals.

Advances in Automotive Radar: A framework on computationally efficient high-resolution frequency estimation

Florian Engels, Philipp Heidenreich, , Friedrich K. Jondral & Markus Wintermantel
This algorithm implements a flexible framework for computationally efficient high-resolution frequency estimation for state-of-the-art automotive radar sensors. By using decoupled frequency estimation in the Fourier domain, high-resolution processing can be applied to either the range, relative velocity, or angular dimension. Real data obtained from series-production automotive radar sensor are supplied to show the effectiveness of the presented framework.

Exploring Causal Relationships in Visual Object Tracking

Karel Lebeda, Simon Hadfield & Richard Bowden
This folder contains source code for causality detection in visual object tracking and for causal predictions, which may be used to help a tracker. It is based on our ICCV2015 paper Exploring Causal Relationships in Visual Object Tracking, our subsequent journal paper, which is currently under review, and parts of my thesis. If you use this code in an academic work, please cite the ICCV paper (unless the follow-up journal has been already published): Lebeda,...

Non-Local Spatial Redundancy Reduction for Bottom-Up Saliency Estimation

, Fei Qi, &
This is a Matlab software algorithm for estimation salience based on redundancy reduction. Please refer to the following paper J. Wu, F. Qi, G. Shi, Y. Lu, Non-local spatial redundancy reduction for bottom-up saliency estimation Journal of Visual Communication and Image Representation, 23(7)(2012), 1158-1166.

MATLAB Code and Data for: Locally Weighted Ensemble Clustering

Dong Huang, Chang-Dong Wang &
In this package, we provide the Matlab source code and experimental data for the ensemble clustering algorithms in our IEEE TCYB paper. If you find the code and data useful for your research, please cite the paper below. Dong Huang, Chang-Dong Wang, Jian-Huang Lai. Locally Weighted Ensemble Clustering. IEEE Transactions on Cybernetics, accepted, 2017. DOI: 10.1109/TCYB.2017.2702343

Code for \"Deriving coral skeletal density from computed tomography (CT): effects of scan and reconstruction settings\"

Thomas DeCarlo
This script computes the mean Hounsfield Units (HU), density, and volume from a computed tomography (CT) scan of a block of Porites coral skeleton as described in DeCarlo (2017). This is the sample scanned with all of the same settings as the "standards", and the CT-derived density accurately matches the density determined from mass a volume. The analysis is performed using the software coralCT (including MATLAB code attached here) from DeCarlo and Cohen (2017) (available...

Incremental Caging Graph Search Algorithm for Equilateral Three-Finger Hands

& Elon D. Rimon
Caging is a robust grasping approach when using multi-fingered robot hands. When an object is caged by the fingers, it cannot escape the hand accidentally. The caging problem is therefore to compute the set of cage formations from which a target immobilizing grasp of the object can be reached, while continuously being caged. Starting at the target immobilizing grasp and gradually opening the hand, the critical cage formation that allows the object to escape the...

Parameter Estimation With Sequential Sampling Using Fisher Information Matrix

Seyed Mohammad Mahdi Alavi, Stefan M. Goetz & Angel V. Peterchev
Sequential parameter estimation (SPE) aims to identify parameters of a system by sequentially taking samples of the system and updating the estimation until a satisfactory level of accuracy is achieved. This package provides you with a tool, based in Matlab, to estimate parameters of a normal function by using SPE based on Fisher information matrix. It also compares the results with estimation based on uniform sampling.

Computer-aided simulation of Stoner–Wohlfarth model: Magnetization process of a single particle.

Zsolt Szabó & Amália Iványi
Magnetization process of a single domain ellipsoidal particle: SW_particle. Developed by: Dr. Zsolt Szabó

Matlab code for \"An Efficient Semi-Lagrangian Algorithm for Simulation of Corona Discharges: The Position-State Separation Method\"

& Marley Becerra
This code was written in MATLAB by Lipeng Liu (lipeng@kth.se) to illustrate how POSS (The Position-State Separation Method) works in the simulation of glow corona discharges. Please refer to the following paper for more information: L. Liu, M. Becerra, "An efficient semi-Lagrangian algorithm for simulation of corona discharges: the position-state separation method, " IEEE Transactions on Plasma Science, volume 44, issue 11 (10 pp), 2016. https://doi.org/10.1109/TPS.2016.2609504

Matlab code for \"Application of the Position-State Separation Method to Simulate Streamer Discharges in Arbitrary Geometries\"

& Marley Becerra
This code was written in MATLAB by Lipeng Liu (lipeng@kth.se) to illustrate how POSS (The Position-State Separation Method) works in the simulation of streamer discharges. Please refer to the following paper for more information: L. Liu, M. Becerra, "Application of the position-state separation method to simulate streamer discharges in arbitrary geometries," IEEE Transactions on Plasma Science, volume 45, issue 4 (9 pp), 2017. https://doi.org/10.1109/TPS.2017.2669330

Probabilistic Least Mean Square Adaptive Filter

Hadi Sadoghi Yazdi & Soheila Ashkezari-T
In the framework of the maximum a posteriori estimation, the present study proposes the probabilistic least mean square (PLMS) adaptive filter for the estimation of an unknown parameter vector from noisy data. PLMS combines parameter space and signal space with combining the prior knowledge of the probability distribution of the process with the evidence existing in the signal. Taking advantage of Kernel density estimation to estimate the prior distribution, PLMS is robust against Gaussian and...

Spectrogram enhancement using multiple window Savitzky Golay (MWSG) filter for robust bird sound detection

, Nisha G Meenakshi & Prasanta Kumar Ghosh
An Unsupervised System for robust bird sound detection using enhanced Multiple Window Savitzky-Golay (MWSG) spectrogram. Proposed method, on average, achieved higher F-Score (10.24% relative) compared to the best of the base line schemes mentioned in the paper

Registration Year

  • 2020
  • 2019
  • 2018
  • 2017

Resource Types

  • Software