Scholars@Duke
Guillermo Sapiro

Guillermo Sapiro

James B. Duke Distinguished Professor Emeritus of Electrical and Computer Engineering
Pierre R. Lamond Department of Electrical and Computer Engineering
Office hours By appointment. Contact via e-mail.  

Scholarly Works - Conferences

The Sensing, Computing, and Devices Opportunities of Computational Behavioral Phenotype: An Autism Spectrum Disorder Case Study

Conference Digest of Technical Papers IEEE International Solid State Circuits Conference · January 1, 2026 The AI revolution will come when high-quality healthcare is accessible to all. We describe steps towards this in autism screening. We developed an app displaying stimuli eliciting behaviors that allow to detect signs of autism. These are automatically dete ... Full text Cite

SSOLE: RETHINKING ORTHOGONAL LOW-RANK EMBEDDING FOR SELF-SUPERVISED LEARNING

Conference 13th International Conference on Learning Representations Iclr 2025 · January 1, 2025 Self-supervised learning (SSL) aims to learn meaningful representations from unlabeled data. Orthogonal Low-rank Embedding (OLE) shows promise for SSL by enhancing intra-class similarity in a low-rank subspace and promoting inter-class dissimilarity in a h ... Cite

Addressing Misspecification in Simulation-based Inference through Data-driven Calibration

Conference Proceedings of Machine Learning Research · January 1, 2025 Driven by steady progress in deep generative modeling, simulation-based inference (SBI) has emerged as the workhorse for inferring the parameters of stochastic simulators. However, recent work has demonstrated that model misspecification can compromise the ... Cite

Digital Phenotyping based on a Mobile App Identifies Distinct and Overlapping Features in Children Diagnosed with Autism versus ADHD

Conference UIST Adjunct 2024 Proceedings of the 37th Annual ACM Symposium on User Interface Software and Technology · October 13, 2024 The high prevalence of autism calls for accessible and scalable technology-assisted screening tools. This will aid in early detection allowing timely access to services and supports. SenseToKnow, a mobile digital phenotyping app, showed potential in elicit ... Full text Cite

Large-scale Validation of a Scalable and Portable Behavioral Digital Screening Tool for Autism at Home

Conference Conference on Human Factors in Computing Systems Proceedings · May 11, 2024 Autism, characterized by challenges in socialization and communication, benefits from early detection for prompt and timely intervention. Traditional autism screening questionnaires often exhibit reduced accuracy in primary care settings and significantly ... Full text Cite

Achieving Group Distributional Robustness and Minimax Group Fairness with Interpolating Classifiers

Conference Proceedings of Machine Learning Research · January 1, 2024 Group distributional robustness optimization methods (GDRO) learn models that guarantee performance across a broad set of demographics. GDRO is often framed as a minimax game where an adversary proposes data distributions under which the model performs poo ... Cite

From Geometry to Causality-Ricci Curvature and the Reliability of Causal Inference on Networks

Conference Proceedings of Machine Learning Research · January 1, 2024 Causal inference on networks faces challenges posed in part by violations of standard identification assumptions due to dependencies between treatment units. Although graph geometry fundamentally influences such dependencies, the potential of geometric too ... Cite

PQ-VAE: Learning Hierarchical Discrete Representations with Progressive Quantization

Conference IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops · January 1, 2024 Variational auto-encoders (VAEs) are widely used in generative modeling and representation learning, with applications ranging from image generation to data compression. However, conventional VAEs face challenges in balancing the tradeoff between compactne ... Full text Cite

Automatic Neurocranial Landmarks Detection from Visible Facial Landmarks Leveraging 3D Head Priors.

Conference Clin Image Based Proced Fairness AI Med Imaging Ethical Philos Issues Med Imaging (2023) · October 2023 The localization and tracking of neurocranial landmarks is essential in modern medical procedures, e.g., transcranial magnetic stimulation (TMS). However, state-of-the-art treatments still rely on the manual identification of head targets and require setti ... Full text Link to item Cite

Minimax Demographic Group Fairness in Federated Learning

Conference ACM International Conference Proceeding Series · June 21, 2022 Federated learning is an increasingly popular paradigm that enables a large number of entities to collaboratively learn better models. In this work, we study minimax group fairness in federated learning scenarios where different participating entities may ... Full text Cite

Using text to teach image retrieval

Conference IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops · June 1, 2021 Image retrieval relies heavily on the quality of the data modeling and the distance measurement in the feature space. Building on the concept of image manifold, we first propose to represent the feature space of images, learned via neural networks, as a gr ... Full text Cite

Nested learning for multi-level classification

Conference ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · January 1, 2021 Deep neural networks models are generally designed and trained for a specific type and quality of data. In this work, we address this problem in the context of nested learning. For many applications, both the input data, at training and testing, and the pr ... Full text Cite

Cirrus: A Long-range Bi-pattern LiDAR Dataset

Conference Proceedings IEEE International Conference on Robotics and Automation · January 1, 2021 In this paper, we introduce Cirrus, a new long-range bi-pattern LiDAR public dataset for autonomous driving tasks such as 3D object detection, critical to highway driving and timely decision making. Our platform is equipped with a high-resolution video cam ... Full text Cite

Blind Pareto Fairness and Subgroup Robustness

Conference Proceedings of Machine Learning Research · January 1, 2021 Much of the work in the field of group fairness addresses disparities between predefined groups based on protected features such as gender, age, and race, which need to be available at train, and often also at test, time. These approaches are static and re ... Cite

Combining multiple contrasts for improving machine learning-based classification of cervical cancers with a low-cost point-of-care Pocket colposcope.

Conference Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference · July 2020 We apply feature-extraction and machine learning methods to multiple sources of contrast (acetic acid, Lugol's iodine and green light) from the white Pocket Colposcope, a low-cost point of care colposcope for cervical cancer screening. We combine features ... Full text Cite

A dictionary approach to domain-invariant learning in deep networks

Conference Advances in Neural Information Processing Systems · January 1, 2020 In this paper, we consider domain-invariant deep learning by explicitly modeling domain shifts with only a small amount of domain-specific parameters in a Convolutional Neural Network (CNN). By exploiting the observation that a convolutional filter can be ... Cite

STOCHASTIC CONDITIONAL GENERATIVE NETWORKS WITH BASIS DECOMPOSITION

Conference 8th International Conference on Learning Representations Iclr 2020 · January 1, 2020 While generative adversarial networks (GANs) have revolutionized machine learning, a number of open questions remain to fully understand them and exploit their power. One of these questions is how to efficiently achieve proper diversity and sampling of the ... Cite

Detecting Adversarial Samples Using Influence Functions and Nearest Neighbors

Conference Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition · January 1, 2020 Deep neural networks (DNNs) are notorious for their vulnerability to adversarial attacks, which are small perturbations added to their input images to mislead their prediction. Detection of adversarial examples is, therefore, a fundamental requirement for ... Full text Cite

Minimax pareto fairness: A multi objective perspective

Conference 37th International Conference on Machine Learning Icml 2020 · January 1, 2020 In this work we formulate and formally characterize group fairness as a multi-objective optimization problem, where each sensitive group risk is a separate objective. We propose a fairness criterion where a classifier achieves minimax risk and is Pareto-ef ... Cite

Instance-based generalization in reinforcement learning

Conference Advances in Neural Information Processing Systems · January 1, 2020 Agents trained via deep reinforcement learning (RL) routinely fail to generalize to unseen environments, even when these share the same underlying dynamics as the training levels. Understanding the generalization properties of RL is one of the challenges o ... Cite

Non-Contact Photoplethysmogram and Instantaneous Heart Rate Estimation from Infrared Face Video

Conference Proceedings International Conference on Image Processing Icip · September 1, 2019 Extracting the instantaneous heart rate (iHR) from face videos has been well studied in recent years. It is well known that changes in skin color due to blood flow can be captured using conventional cameras. One of the main limitations of methods that rely ... Full text Cite

Adversarially Learned Representations for Information Obfuscation and Inference

Conference Proceedings of Machine Learning Research · January 1, 2019 Data collection and sharing are pervasive aspects of modern society. This process can either be voluntary, as in the case of a person taking a facial image to unlock his/her phone, or incidental, such as traffic cameras collecting videos on pedestrians. An ... Cite

Adversarially learned representations for information obfuscation and inference

Conference 36th International Conference on Machine Learning Icml 2019 · January 1, 2019 Data collection and sharing are pervasive aspects of modern society. This process can either be voluntary, as in the case of a person taking a facial image to unlock his/her phone, or incidental, such as traffic cameras collecting videos on pedestrians. An ... Cite

Classifying Pump-Probe Images of Melanocytic Lesions Using the WEYL Transform

Conference ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · September 10, 2018 Diagnosis of melanoma is fraught with uncertainty, and discordance rates among physicians remain high because of the lack of a definitive criterion. Motivated by this challenge, this paper first introduces the Patch Weyl transform (PWT), a 2-dimensional va ... Full text Cite

The Learned Inexact Project Gradient Descent Algorithm

Conference ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · September 10, 2018 Accelerating iterative algorithms for solving inverse problems using neural networks have become a very popular strategy in the recent years. In this work, we propose a theoretical analysis that may provide an explanation for its success. Our theory relies ... Full text Cite

Intelligent synthesis driven model calibration: framework and face recognition application

Conference Proceedings 2017 IEEE International Conference on Computer Vision Workshops Iccvw 2017 · January 19, 2018 Deep Neural Networks (DNNs) that achieve state-of-the-art results are still prone to suffer performance degradation when deployed in many real-world scenarios due to shifts between the training and deployment domains. Limited data from a given setting can ... Full text Cite

Learning to identify while failing to discriminate

Conference Proceedings 2017 IEEE International Conference on Computer Vision Workshops Iccvw 2017 · January 19, 2018 Privacy and fairness are critical in computer vision applications, in particular when dealing with human identification. Achieving a universally secure, private, and fair systems is practically impossible as the exploitation of additional data can reveal p ... Full text Cite

Image processing and machine learning techniques to automate diagnosis of Lugol's iodine cervigrams for a low-cost point-of-care digital colposcope

Conference Progress in Biomedical Optics and Imaging Proceedings of SPIE · January 1, 2018 The world health organization recommends visual inspection with acetic acid (VIA) and/or Lugol's Iodine (VILI) for cervical cancer screening in low-resource settings. Human interpretation of diagnostic indicators for visual inspection is qualitative, subje ... Full text Cite

DCFNet: Deep Neural Network with Decomposed Convolutional Filters

Conference Proceedings of Machine Learning Research · January 1, 2018 Filters in a Convolutional Neural Network (CNN) contain model parameters learned from enormous amounts of data. In this paper, we suggest to decompose convolutional filters in CNN as a truncated expansion with pre-fixed bases, namely the Decomposed Convolu ... Cite

Deep video deblurring for hand-held cameras

Conference Proceedings 30th IEEE Conference on Computer Vision and Pattern Recognition Cvpr 2017 · November 6, 2017 Motion blur from camera shake is a major problem in videos captured by hand-held devices. Unlike single-image deblurring, video-based approaches can take advantage of the abundant information that exists across neighboring frames. As a result the best perf ... Full text Cite

Nonnegative matrix underapproximation for robust multiple model fitting

Conference Proceedings 30th IEEE Conference on Computer Vision and Pattern Recognition Cvpr 2017 · November 6, 2017 In this work, we introduce a highly efficient algorithm to address the nonnegative matrix underapproximation (NMU) problem, i.e., nonnegative matrix factorization (NMF) with an additional underapproximation constraint. NMU results are interesting as, compa ... Full text Cite

A Sparse Bayesian Learning Algorithm for White Matter Parameter Estimation from Compressed Multi-shell Diffusion MRI.

Conference Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention · September 2017 We propose a sparse Bayesian learning algorithm for improved estimation of white matter fiber parameters from compressed (under-sampled q-space) multi-shell diffusion MRI data. The multi-shell data is represented in a dictionary form using a non-monoexpone ... Full text Cite

Generalization error of deep neural networks: Role of classification margin and data structure

Conference 2017 12th International Conference on Sampling Theory and Applications Sampta 2017 · September 1, 2017 Understanding the generalization properties of deep learning models is critical for their successful usage in many applications, especially in the regimes where the number of training samples is limited. We study the generalization properties of deep neura ... Full text Cite

RealSense = real heart rate: Illumination invariant heart rate estimation from videos

Conference 2016 6th International Conference on Image Processing Theory Tools and Applications Ipta 2016 · January 17, 2017 Recent studies validated the feasibility of estimating heart rate from human faces in RGB video. However, test subjects are often recorded under controlled conditions, as illumination variations significantly affect the RGB-based heart rate estimation accu ... Full text Cite

Generalization error of invariant classifiers

Conference Proceedings of the 20th International Conference on Artificial Intelligence and Statistics, AISTATS 2017 · January 1, 2017 Copyright 2017 by the author(s). This paper studies the generalization error of invariant classifiers. In particular, we consider the common scenario where the classification task is invariant to certain transformations of the input, and that the classifie ... Cite

Generalization error of invariant classifiers

Conference Proceedings of the 20th International Conference on Artificial Intelligence and Statistics Aistats 2017 · January 1, 2017 This paper studies the generalization error of invariant classifiers. In particular, we consider the common scenario where the classification task is invariant to certain transformations of the input, and that the classifier is constructed (or learned) to ... Cite

Tell me where you are and i tell you where you are going: Estimation of dynamic mobility graphs

Conference Proceedings of the IEEE Sensor Array and Multichannel Signal Processing Workshop · September 15, 2016 The interest in problems related to graph inference has been increasing significantly during the last decade. However, the vast majority of the problems addressed are either static, or systems where changes in one node are immediately reflected in other no ... Full text Cite

A short-graph fourier transform via personalized pagerank vectors

Conference ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · May 18, 2016 The short-time Fourier transform (STFT) is widely used to analyze the spectra of temporal signals that vary through time. Signals defined over graphs, due to their intrinsic complexity, exhibit large variations in their patterns. In this work we propose a ... Full text Cite

Synthesis-based low-cost gaze analysis

Conference Communications in Computer and Information Science · January 1, 2016 Gaze analysis has gained much popularity over the years due to its relevance in a wide array of applications, including humancomputer interaction, fatigue detection, and clinical mental health diagnosis. However, accurate gaze estimation from low resolutio ... Full text Cite

Clinical deep brain stimulation region prediction using regression forests from high-field MRI

Conference Proceedings International Conference on Image Processing Icip · December 9, 2015 This paper presents a prediction framework of brain subcortical structures which are invisible on clinical low-field MRI, learning detailed information from ultrahigh-field MR training data. Volumetric segmentation of Deep Brain Stimulation (DBS) structure ... Full text Cite

Multi-temporal foreground detection in videos

Conference Proceedings International Conference on Image Processing Icip · December 9, 2015 A common task in video processing is the binary separation of a video's content into either background or moving foreground. However, many situations require a foreground analysis with a finer temporal granularity, in particular for objects or people which ... Full text Cite

Cross-modality pose-invariant facial expression

Conference Proceedings International Conference on Image Processing Icip · December 9, 2015 In this work, we present a dictionary learning based framework for robust, cross-modality, and pose-invariant facial expression recognition. The proposed framework first learns a dictionary that i) contains both 3D shape and morphological information as we ... Full text Cite

Burst deblurring: Removing camera shake through fourier burst accumulation

Conference Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition · October 14, 2015 Numerous recent approaches attempt to remove image blur due to camera shake, either with one or multiple input images, by explicitly solving an inverse and inherently ill-posed deconvolution problem. If the photographer takes a burst of images, a modality ... Full text Cite

Alignment with intra-class structure can improve classification

Conference ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings · August 4, 2015 High dimensional data is modeled using low-rank subspaces, and the probability of misclassification is expressed in terms of the principal angles between subspaces. The form taken by this expression motivates the design of a new feature extraction method t ... Full text Cite

Clinical subthalamic nucleus prediction from high-field brain MRI

Conference Proceedings International Symposium on Biomedical Imaging · July 21, 2015 The subthalamic nucleus (STN) within the sub-cortical region of the Basal ganglia is a crucial targeting structure for Parkinson's Deep brain stimulation (DBS) surgery. Volumetric segmentation of such small and complex structure, which is elusive in clinic ... Full text Cite

Geometry-aware deep transform

Conference Proceedings of the IEEE International Conference on Computer Vision · February 17, 2015 Many recent efforts have been devoted to designing sophisticated deep learning structures, obtaining revolutionary results on benchmark datasets. The success of these deep learning methods mostly relies on an enormous volume of labeled training samples to ... Full text Cite

Robust prediction of clinical deep brain stimulation target structures via the estimation of influential high-field MR atlases

Conference Lecture Notes in Computer Science Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics · January 1, 2015 This work introduces a robust framework for predicting Deep Brain Stimulation (DBS) target structures which are not identifiable on standard clinical MRI. While recent high-field MR imaging allows clear visualization of DBS target structures, such high-fie ... Full text Cite

Intel realsense= real low cost gaze

Conference Image Processing (ICIP), 2015 IEEE International Conference on · 2015 Cite

Discriminative robust transformation learning

Conference Advances in Neural Information Processing Systems · January 1, 2015 This paper proposes a framework for learning features that are robust to data variation, which is particularly important when only a limited number of training samples are available. The framework makes it possible to tradeoff the discriminative value of l ... Cite

From local to global communities in large networks through consensus

Conference Lecture Notes in Computer Science Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics · January 1, 2015 Given a universe of local communities of a large network, we aim at identifying the meaningful and consistent communities in it. We address this from a new perspective as the process of obtaining consensual community detections and formalize it as a bi-clu ... Full text Cite

Low-cost Gaze and Pulse Analysis using RealSense

Conference Proceedings of the 5th EAI International Conference on Wireless Mobile Communication and Healthcare · 2015 Cite

Temporal compressive sensing for video

Conference · January 1, 2015 Video camera architects must design cameras capable of high-quality, dynamic event capture, while adhering to power and communications constraints. Though modern imagers are capable of both simultaneous spatial and temporal resolutions at micrometer and mi ... Full text Cite

Data representation using the Weyl transform

Conference 3rd International Conference on Learning Representations, ICLR 2015 - Workshop Track Proceedings · January 1, 2015 © 2015 International Conference on Learning Representations, ICLR. All rights reserved. The Weyl transform is introduced as a rich framework for data representation. Transform coefficients are connected to the Walsh-Hadamard transform of multiscale autocor ... Cite

On the stability of deep networks

Conference 3rd International Conference on Learning Representations, ICLR 2015 - Workshop Track Proceedings · January 1, 2015 © 2015 International Conference on Learning Representations, ICLR. All rights reserved. In this work we study the properties of deep neural networks (DNN) with random weights. We formally prove that these networks perform a distance-preserving embedding of ... Cite

Data representation using the Weyl transform

Conference 3rd International Conference on Learning Representations Iclr 2015 Workshop Track Proceedings · January 1, 2015 The Weyl transform is introduced as a rich framework for data representation. Transform coefficients are connected to the Walsh-Hadamard transform of multiscale autocorrelations, and different forms of dyadic periodicity in a signal are shown to appear as ... Cite

Random forests can hash

Conference 3rd International Conference on Learning Representations Iclr 2015 Workshop Track Proceedings · January 1, 2015 Hash codes are a very efficient data representation needed to be able to cope with the ever growing amounts of data. We introduce a random forest semantic hashing scheme with information-theoretic code aggregation, showing for the first time how random for ... Cite

On the stability of deep networks

Conference 3rd International Conference on Learning Representations Iclr 2015 Workshop Track Proceedings · January 1, 2015 In this work we study the properties of deep neural networks (DNN) with random weights. We formally prove that these networks perform a distance-preserving embedding of the data. Based on this we then draw conclusions on the size of the training data and t ... Cite

Low-Rank Spatio-Temporal Video Segmentation

Conference 26th British Machine Vision Conference Bmvc 2015 · January 1, 2015 Robust Principal Component Analysis (RPCA) has generated a great amount of interest for background/foreground estimation in videos. The central hypothesis in this setting is that a video's background can be well-represented by a low-rank model. However, in ... Full text Cite

Low-cost compressive sensing for color video and depth

Conference Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition · September 24, 2014 A simple and inexpensive (low-power and low-bandwidth) modification is made to a conventional off-the-shelf color video camera, from which we recover multiple color frames for each of the original measured frames, and each of the recovered frames can be fo ... Full text Cite

Learning compressed image classification features

Conference 2014 IEEE International Conference on Image Processing Icip 2014 · January 28, 2014 Learning a transformation-based dimension reduction, thereby compressive, technique for classification is here proposed. High-dimensional data often approximately lie in a union of low-dimensional subspaces. We propose to perform dimension reduction by lea ... Full text Cite

Learning Transformations

Conference 2014 IEEE International Conference on Image Processing Icip 2014 · January 28, 2014 A low-rank transformation learning framework for subspace clustering and classification is here proposed. Many high-dimensional data, such as face images and motion sequences, approximately lie in a union of low-dimensional subspaces. The corresponding sub ... Full text Cite

Intersecting 2D lines: A simple method for detecting vanishing points

Conference 2014 IEEE International Conference on Image Processing Icip 2014 · January 28, 2014 We present a simple and powerful technique for testing with a prescribed precision whether a set of 2D lines meet at a given point. The method is based on a probabilistic framework and has a fundamental geometric interpretation. We use this technique for d ... Full text Cite

Rich club analysis of structural brain connectivity at 7 tesla versus 3 tesla

Conference Mathematics and Visualization · January 1, 2014 The ‘rich club’ is a relatively new concept in brain connectivity analysis, which identifies a core of densely interconnected high-degree nodes. Establishing normative measures for rich club organization is vital, as is understanding how scanning parameter ... Full text Cite

Sparse similarity-preserving hashing

Conference 2nd International Conference on Learning Representations, ICLR 2014 - Conference Track Proceedings · January 1, 2014 © 2014 International Conference on Learning Representations, ICLR. All rights reserved. In recent years, a lot of attention has been devoted to efficient nearest neighbor search by means of similarity-preserving hashing. One of the plights of existing hash ... Cite

The work of Stanley Osher

Conference Proceeding of the International Congress of Mathematicans Icm 2014 · January 1, 2014 In this paper we briefly present some of Stanley Osher's contributions in the areas of high resolution shock capturing methods, level set methods, partial differential equation (PDE) based methods in computer vision and image processing, and optimization. ... Cite

Sparse similarity-preserving hashing

Conference 2nd International Conference on Learning Representations Iclr 2014 Conference Track Proceedings · January 1, 2014 In recent years, a lot of attention has been devoted to efficient nearest neighbor search by means of similarity-preserving hashing. One of the plights of existing hashing techniques is the intrinsic trade-off between performance and computational complexi ... Cite

Learning transformations for classification forests

Conference 2nd International Conference on Learning Representations Iclr 2014 Conference Track Proceedings · January 1, 2014 This work introduces a transformation-based learner model for classification forests. The weak learner at each split node plays a crucial role in a classification tree. We propose to optimize the splitting objective by learning a linear transformation on s ... Cite

Reflective Symmetry Detection by Rectifying Randomized Correspondences

Conference Bmvc 2013 Electronic Proceedings of the British Machine Vision Conference 2013 · January 1, 2013 We present a method for detecting bilateral or reflective symmetries in images. We pose the problem as an instance of a multiple model estimation problem. We build candidate symmetry models by randomly sampling minimal sets of SIFT matches. Since these sym ... Full text Cite

Bilevel sparse models for polyphonic music transcription

Conference Proceedings of the 14th International Society for Music Information Retrieval Conference Ismir 2013 · January 1, 2013 In this work, we propose a trainable sparse model for automatic polyphonic music transcription, which incorporates several successful approaches into a unified optimization framework. Our model combines unsupervised synthesis models similar to latent compo ... Cite

Compressive sensing for video using a passive coding element

Conference Optics Infobase Conference Papers · January 1, 2013 We present a prototype system that utilizes mechanical translation of a passive coding element to compress high-speed temporal information into low-framerate video sequences. Reconstructions of 148 frames per experimental coded snapshot are reported. © OSA ... Full text Cite

A computer vision approach for the assessment of autism-related behavioral markers

Conference 2012 IEEE International Conference on Development and Learning and Epigenetic Robotics ICDL 2012 · December 1, 2012 The early detection of developmental disorders is key to child outcome, allowing interventions to be initiated that promote development and improve prognosis. Research on autism spectrum disorder (ASD) suggests behavioral markers can be observed late in th ... Full text Cite

Detecting risk-markers in children in a preschool classroom

Conference IEEE International Conference on Intelligent Robots and Systems · December 1, 2012 Early intervention in mental disorders can dramatically increase an individual's quality of life. Additionally, when symptoms of mental illness appear in childhood or adolescence, they represent the later stages of a process that began years earlier. One g ... Full text Cite

Kernelized probabilistic matrix factorization: Exploiting graphs and side information

Conference Proceedings of the 12th SIAM International Conference on Data Mining Sdm 2012 · January 1, 2012 We propose a new matrix completion algorithm| Kernelized Probabilistic Matrix Factorization (KPMF), which effectively incorporates external side information into the matrix factorization process. Unlike Probabilistic Matrix Factorization (PMF) [14], which ... Full text Cite

Sparse modeling for hyperspectral imagery with LiDAR data fusion for subpixel mapping

Conference International Geoscience and Remote Sensing Symposium IGARSS · January 1, 2012 Several studies suggest that the use of geometric features along with spectral information improves the classification and visualization quality of hyperspectral imagery. These studies normally make use of spatial neighborhoods of hyperspectral pixels for ... Full text Cite

Real-time online singing voice separation from monaural recordings using robust low-rank modeling

Conference Proceedings of the 13th International Society for Music Information Retrieval Conference Ismir 2012 · January 1, 2012 Separating the leading vocals from the musical accompaniment is a challenging task that appears naturally in several music processing applications. Robust principal component analysis (RPCA) has been recently employed to this problem producing very success ... Cite

Seeing 3D Objects in a Single 2D Image

Conference Proceedings of the IEEE International Conference on Computer Vision · January 1, 2009 A general framework simultaneously addressing pose estimation, 2D segmentation, object recognition, and 3D reconstruction from a single image is introduced in this paper. The proposed approach partitions 3D space into voxels and estimates the voxel states ... Full text Cite

Supervised Dictionary Learning.

Conference NIPS · 2008 Cite

GENERALIZED NEWTON METHODS FOR ENERGY FORMULATIONS IN IMAGE PROCESSING

Conference 2008 IEEE International Conference on Image Processing, Proceedings · 2008 Link to item Cite

What Can Casual Walkers Tell Us About A 3D Scene?

Conference 2007 IEEE 11th International Conference on Computer Vision · 2007 Full text Cite

A graph-based foreground representation and its application in example based people matching in video

Conference 2007 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, VOLS 1-7 · 2007 Link to item Cite

Distancecut: Interactive segmentation and matting of images and videos

Conference 2007 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, VOLS 1-7 · 2007 Link to item Cite

Multiscale sparse image representation with learned dictionaries

Conference 2007 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, VOLS 1-7 · 2007 Link to item Cite

Stratification Learning: Detecting Mixed Density and Dimensionality in High Dimensional Point Clouds

Conference Nips 2006 Proceedings of the 19th International Conference on Neural Information Processing Systems · January 1, 2006 The study of point cloud data sampled from a stratification, a collection of manifolds with possible different dimensions, is pursued in this paper. We present a technique for simultaneously soft clustering and estimating the mixed dimensionality and densi ... Cite

Video inpainting of occluding and occluded objects

Conference 2005 International Conference on Image Processing (ICIP), Vols 1-5 · 2005 Link to item Cite

Tracking of moving objects under severe and total occlusions

Conference 2005 International Conference on Image Processing (ICIP), Vols 1-5 · January 1, 2005 Link to item Cite

Inpainting the colors

Conference 2005 International Conference on Image Processing (ICIP), Vols 1-5 · 2005 Link to item Cite

Level set and PDE methods for computer graphics

Conference ACM SIGGRAPH 2004 Course Notes SIGGRAPH 2004 · August 8, 2004 Level set methods, an important class of partial differential equation (PDE) methods, define dynamic surfaces implicitly as the level set (iso-surface) of a sampled, evolving nD function. The course begins with preparatory material that introduces the conc ... Full text Cite

Morse description and geometric encoding of digital elevation maps

Conference FREE BOUNDARY PROBLEMS: THEORY AND APPLICATIONS · 2004 Link to item Cite

Harmonic map flows and image processing

Conference FOUNDATIONS OF COMPUTATIONAL MATHEMATICS · 2001 Link to item Cite

Variational problems and PDEs on implicit surfaces

Conference Proceedings IEEE Workshop on Variational and Level Set Methods in Computer Vision Vlsm 2001 · January 1, 2001 A novel framework for solving variational problems and partial differential equations for scalar and vector-valued data defined on surfaces is introduced. The key idea is to implicitly represent the surface as the level set of a higher dimensional function ... Full text Cite

Noise-resistant affine skeletons of planar curves

Conference Lecture Notes in Computer Science Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics · January 1, 2000 A new definition of affine invariant skeletons for shape re- presentation is introduced. A point belongs to the affine skeleton if and only if it is equidistant from at least two points of the curve, with the distance being a minima and given by the areas ... Full text Cite

Affine versions of the symmetry set

Conference REAL AND COMPLEX SINGULARITIES · 2000 Link to item Cite

A windows-based user friendly system for image analysis with partial differential equations

Conference Lecture Notes in Computer Science Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics · January 1, 1999 In this paper we present and briefly describe a Windows user-friendly system designed to assist with the analysis of images in general, and biomedical images in particular. The system, which is being made publicly available to the research community, imple ... Full text Cite

Morphing active contours

Conference Lecture Notes in Computer Science Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics · January 1, 1999 A method for deforming curves in a given image to a desired position in a second image is introduced in this paper. The algorithm is based on deforming the first image toward the second one via a partial differential equation, while tracking the deformatio ... Full text Cite

Region tracking on surfaces deforming via level-sets methods

Conference Lecture Notes in Computer Science Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics · January 1, 1999 Since the work by Osher and Sethian on level-sets algorithms for numerical shape evolutions, this technique has been used for a large number of applications in numerous fields. In medical imaging, this numerical technique has been successfully used for exa ... Full text Cite

Edges as outliers: Anisotropic smoothing using local image statistics

Conference Lecture Notes in Computer Science Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics · January 1, 1999 Edges are viewed as statistical outliers with respect to local image gradient magnitudes. Within local image regions we compute a robust statistical measure of the gradient variation and use this in an anisotropic diffusion framework to determine a spatial ... Full text Cite

Anisotropic smoothing of posterior probabilities

Conference DYNAMICAL SYSTEMS, CONTROL, CODING, COMPUTER VISION · 1999 Link to item Cite

Robust anisotropic diffusion: Connections between robust statistics, line processing, and anisotropic diffusion

Conference Lecture Notes in Computer Science Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics · January 1, 1997 Relations between anisotropic diffusion and robust statistics are described in this paper. We show that anisotropic diffusion can be seen as a robust estimation procedure that estimates a piecewise smooth image from a noisy input image. The "edge-stopping" ... Full text Cite

Three dimensional object modeling via minimal surfaces

Conference Lecture Notes in Computer Science · January 1, 1996 A novel geometric approach for 3D object segmentation and representation is presented. The scheme is based on geometric deformable surfaces moving towards the objects to be detected. We show that this model is equivalent to the computation of surfaces of m ... Full text Cite

An image evolution approach for contrast enhancement

Conference Proceedings of SPIE the International Society for Optical Engineering · September 1, 1995 An algorithm for histogram modification via image evolution equations is first presented in this paper. We show that the image histogram can be modified to achieve any given distribution as the steady state solution of this partial differential equation. W ... Full text Cite

Simultaneous contrast improvement and denoising via diffusion related equations

Conference Proceedings of SPIE the International Society for Optical Engineering · August 11, 1995 The explicit use of partial differential equations (PDE's) in image processing became a major topic of study in the last years. In this work we present an algorithm for histogram modification via PDE's. We show that the histogram can be modified to achieve ... Full text Cite

Object detection and measurements in medical images via geodesic deformable contours

Conference Proceedings of SPIE the International Society for Optical Engineering · August 11, 1995 A novel scheme for performing detection and measurements in medical images is presented. The technique is based on active contours evolving in time according to intrinsic geometric measures of the image. The evolving contours naturally split and merge, all ... Full text Cite

Geometric invariant signatures and flows: Classification and applications in image analysis

Conference Proceedings of SPIE the International Society for Optical Engineering · October 25, 1994 Based on modern invariant theory and symmetry groups, a high level way of defining invariant geometricflows for a given Lie group is described in this work. We then analyze in more detail different subgroups ofthe projective group, which are of special int ... Full text Cite

Experiments on geometric image enhancement

Conference Proceedings International Conference on Image Processing Icip · January 1, 1994 In this paper we experiments with geometric algorithms for image smoothing. Examples are given for MRI and ATR data. We emphasize experiments with the affine invariant geometric smoother or affine heat equation, originally developed for binary shape smooth ... Full text Cite

Area and length preserving geometric invariant scale-spaces

Conference Lecture Notes in Computer Science Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics · January 1, 1994 In this paper, area preserving geometric multi-scale representations of planar curves are described. This allows geometric smoothing without shrinkage at the same time preserving all the scale-space properties. The representations are obtained deforming th ... Full text Cite

The ubiquitous ellipse

Conference CURVES AND SURFACES IN GEOMETRIC DESIGN · 1994 Link to item Cite

IMAGE SMOOTHING BASED ON AN AFFINE INVARIANT CURVE FLOW

Conference PROCEEDINGS OF THE TWENTY-SEVENTH ANNUAL CONFERENCE ON INFORMATION SCIENCES AND SYSTEMS · 1993 Link to item Cite

Morphological image coding via bit-plane decomposition and a new skeleton representation

Conference Proceedings 17th Convention of Electrical and Electronics Engineers in Israel Eeis 1991 · January 1, 1991 A new approach for image coding based on bit-plane decomposition and binary morphological operations is presented. The image is first processed by an errordiffusion technique in order to reduce the number of bitplanes without a significant quality degradat ... Full text Cite