Search For:

Displaying 1-48 out of 48 total
Scalable Discriminant Feature Selection for Image Retrieval and Recognition
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Nuno Vasconcelos, Manuela Vasconcelos
Issue Date:July 2004
pp. 770-775
Problems such as object recognition or image retrieval require feature selection (FS) algorithms that scale well enough to be applicable to databases containing large numbers of image classes and large amounts of data per class. We exploit recent connectio...
 
High Detection-rate Cascades for Real-Time Object Detection
Found in: Computer Vision, IEEE International Conference on
By Hamed Masnadi-Shirazi, Nuno Vasconcelos
Issue Date:October 2007
pp. 1-6
A new strategy is proposed for the design of cascaded object detectors of high detection-rate. The problem of jointly minimizing the false-positive rate and classification complexity of a cascade, given a constraint on its detection rate, is considered. It...
 
Bottom-up saliency is a discriminant process
Found in: Computer Vision, IEEE International Conference on
By Dashan Gao, Nuno Vasconcelos
Issue Date:October 2007
pp. 1-6
A bottom-up visual saliency detector is proposed, following a decision-theoretic formulation of saliency, previously developed for top-down processing (object recognition) [5]. The saliency of a given location of the visual field is defined as the power of...
 
Classifying Video with Kernel Dynamic Textures
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Antoni B. Chan, Nuno Vasconcelos
Issue Date:June 2007
pp. 1-6
The dynamic texture is a stochastic video model that treats the video as a sample from a linear dynamical system. The simple model has been shown to be surprisingly useful in domains such as video synthesis, video segmentation, and video classification. Ho...
 
Discriminant Interest Points are Stable
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Dashan Gao, Nuno Vasconcelos
Issue Date:June 2007
pp. 1-6
A study of the performance of recently introduced discriminant methods for interest point detection [6, 14] is presented. It has been previously shown that the resulting interest points are more informative for object recognition than those produced by the...
 
Natural Image Statistics and Low-Complexity Feature Selection
Found in: IEEE Transactions on Pattern Analysis and Machine Intelligence
By Manuela Vasconcelos, Nuno Vasconcelos
Issue Date:February 2009
pp. 228-244
Low-complexity feature selection is analyzed in the context of visual recognition. It is hypothesized that high-order dependences of bandpass features contain little information for discrimination of natural images. This hypothesis is characterized formall...
 
Weakly Supervised Top-down Image Segmentation
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Manuela Vasconcelos, Nuno Vasconcelos, Gustavo Carneiro
Issue Date:June 2006
pp. 1001-1006
There has recently been significant interest in top-down image segmentation methods, which incorporate the recognition of visual concepts as an intermediate step of segmentation. This work addresses the problem of top-down segmentation with weak supervisio...
 
Class-Specific Simplex-Latent Dirichlet Allocation for Image Classification
Found in: 2013 IEEE International Conference on Computer Vision (ICCV)
By Mandar Dixit,Nikhil Rasiwasia,Nuno Vasconcelos
Issue Date:December 2013
pp. 2672-2679
An extension of the latent Dirichlet allocation (LDA), denoted class-specific-simplex LDA (css-LDA), is proposed for image classification. An analysis of the supervised LDA models currently used for this task shows that the impact of class information on t...
 
Motion vector refinement for FRUC using saliency and segmentation
Found in: Multimedia and Expo, IEEE International Conference on
By Natan Jacobson, Yen-Lin Lee, Vijay Mahadevan, Nuno Vasconcelos, Truong Q. Nguyen
Issue Date:July 2010
pp. 778-783
Motion-Compensated Frame Interpolation (MCFI) is a technique used extensively for increasing the temporal frequency of a video sequence. In order to obtain a high quality interpolation, the motion field between frames must be well-estimated. However, many ...
 
On the design of robust classifiers for computer vision
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Hamed Masnadi-Shirazi, Vijay Mahadevan, Nuno Vasconcelos
Issue Date:June 2010
pp. 779-786
The design of robust classifiers, which can contend with the noisy and outlier ridden datasets typical of computer vision, is studied. It is argued that such robustness requires loss functions that penalize both large positive and negative margins. The pro...
 
Anomaly detection in crowded scenes
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Vijay Mahadevan, Weixin Li, Viral Bhalodia, Nuno Vasconcelos
Issue Date:June 2010
pp. 1975-1981
A novel framework for anomaly detection in crowded scenes is presented. Three properties are identified as important for the design of a localized video representation suitable for anomaly detection in such scenes: 1) joint modeling of appearance and dynam...
 
A study of query by semantic example
Found in: Computer Vision and Pattern Recognition Workshop
By Nikhil Rasiwasia, Nuno Vasconcelos
Issue Date:June 2008
pp. 1-8
In recent years, query-by-semantic-example (QBSE) has become a popular approach to do content based image retrieval [20, 23, 18]. QBSE extends the well established query-by-example retrieval paradigm to the semantic domain. While various authors have point...
 
Privacy preserving crowd monitoring: Counting people without people models or tracking
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Antoni B. Chan, Zhang-Sheng John Liang, Nuno Vasconcelos
Issue Date:June 2008
pp. 1-7
We present a privacy-preserving system for estimating the size of inhomogeneous crowds, composed of pedestrians that travel in different directions, without using explicit object segmentation or tracking. First, the crowd is segmented into components of ho...
 
Background subtraction in highly dynamic scenes
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Vijay Mahadevan, Nuno Vasconcelos
Issue Date:June 2008
pp. 1-6
A new algorithm is proposed for background subtraction in highly dynamic scenes. Background subtraction is equated to the dual problem of saliency detection: background points are those considered not salient by suitable comparison of object and background...
 
Scene classification with low-dimensional semantic spaces and weak supervision
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Nikhil Rasiwasia, Nuno Vasconcelos
Issue Date:June 2008
pp. 1-6
A novel approach to scene categorization is proposed. Similar to previous works of [11, 15, 3, 12], we introduce an intermediate space, based on a low dimensional semantic “theme” image representation. However, instead of learning the themes in an unsuperv...
 
Object-Based Regions of Interest for Image Compression
Found in: Data Compression Conference
By Sunhyoung Han, Nuno Vasconcelos
Issue Date:March 2008
pp. 132-141
A fully automated architecture for object-based region of interest (ROI) detection is proposed. ROI's are defined as regions containing user defined objects of interest, and an efficient algorithm is developed for the detection of such regions. The algorit...
 
From Pixels to Semantic Spaces: Advances in Content-Based Image Retrieval
Found in: Computer
By Nuno Vasconcelos
Issue Date:July 2007
pp. 20-26
The image retrieval paradigm has evolved from low-level image representations to semantic concept models to higher-level semantic inferences. UCSD's Statistical Visual Computing Laboratory has developed effective techniques for each paradigm that equate re...
 
Mixtures of Dynamic Textures
Found in: Computer Vision, IEEE International Conference on
By Antoni B. Chan, Nuno Vasconcelos
Issue Date:October 2005
pp. 641-647
A dynamic texture is a linear dynamical system used to model a single video as a sample from a spatio-temporal stochastic process. In this work, we introduce the mixture of dynamic textures, which models a collection of videos consisting of different visua...
 
An Experimental Comparison of Three Guiding Principles for the Detection Salient Image Locations: Stability, Complexity, and Discrimination
Found in: Computer Vision and Pattern Recognition Workshop
By Dashan Gao, Nuno Vasconcelos
Issue Date:June 2005
pp. 84
<p>We present an experimental comparison of the performance of representative saliency detectors from three guiding principles for the detection of salient image locations: locations of maximum stability with respect to image transformations, locatio...
 
Integrated Learning of Saliency, Complex Features, and Object Detectors from Cluttered Scenes
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Dashan Gao, Nuno Vasconcelos
Issue Date:June 2005
pp. 282-287
A novel procedure for object detection from cluttered scenes is proposed. It consists of an integrated solution to the problems of learning 1) a saliency detection module tuned to a class of objects of interest, 2) a set of complex features that achieves t...
 
Minimum Bayes Error Features for Visual Recognition by Sequential Feature Selection and Extraction
Found in: Computer and Robot Vision, Canadian Conference
By Gustavo Carneiro, Nuno Vasconcelos
Issue Date:May 2005
pp. 253-260
The extraction of optimal features, in a classification sense, is still quite challenging in the context of large-scale classification problems (such as visual recognition), involving a large number of classes and significant amounts of training data per c...
 
Feature Selection by Maximum Marginal Diversity: optimality and implications for visual recognition
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Nuno Vasconcelos
Issue Date:June 2003
pp. 762
We have recently shown that 1) the infomax principle for the organization of perceptual systems leads to visual recognition architectures that are nearly optimal in the minimum Bayes error sense, and 2) a quantity which plays an important role in infomax s...
 
Empirical Bayesian EM-based Motion Segmentation
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Nuno Vasconcelos, Andrew Lippman
Issue Date:June 1997
pp. 527
A recent trend in motion-based segmentation has been to rely on statistical procedures derived from Expectation-Maximization (EM) principles. EM-based approaches have various attractives for segmentation, such as proceeding by taking non-greedy soft decisi...
 
Library-Based Coding: A Representation for Efficient Video Compression and Retrieval
Found in: Data Compression Conference
By Nuno Vasconcelos, Andrew Lippman
Issue Date:March 1997
pp. 121
The ubiquity of networking and computational capacity associated with the new communications media unveil a universe of new requirements for image representation. Among such requirements is the ability of the representation used for coding to support highe...
 
Dynamic Pooling for Complex Event Recognition
Found in: 2013 IEEE International Conference on Computer Vision (ICCV)
By Weixin Li,Qian Yu,Ajay Divakaran,Nuno Vasconcelos
Issue Date:December 2013
pp. 2728-2735
The problem of adaptively selecting pooling regions for the classification of complex video events is considered. Complex events are defined as events composed of several characteristic behaviors, whose temporal configuration can change from sequence to se...
 
Learning Optimal Embedded Cascades
Found in: IEEE Transactions on Pattern Analysis and Machine Intelligence
By Mohammad Javad Saberian,Nuno Vasconcelos
Issue Date:October 2012
pp. 2005-2018
The problem of automatic and optimal design of embedded object detector cascades is considered. Two main challenges are identified: optimization of the cascade configuration and optimization of individual cascade stages, so as to achieve the best tradeoff ...
 
Holistic Context Models for Visual Recognition
Found in: IEEE Transactions on Pattern Analysis and Machine Intelligence
By Nikhil Rasiwasia,Nuno Vasconcelos
Issue Date:May 2012
pp. 902-917
A novel framework to context modeling based on the probability of co-occurrence of objects and scenes is proposed. The modeling is quite simple, and builds upon the availability of robust appearance classifiers. Images are represented by their posterior pr...
 
Cost-Sensitive Boosting
Found in: IEEE Transactions on Pattern Analysis and Machine Intelligence
By Hamed Masnadi-Shirazi, Nuno Vasconcelos
Issue Date:February 2011
pp. 294-309
A novel framework is proposed for the design of cost-sensitive boosting algorithms. The framework is based on the identification of two necessary conditions for optimal cost-sensitive learning that 1) expected losses must be minimized by optimal cost-sensi...
 
Spatiotemporal Saliency in Dynamic Scenes
Found in: IEEE Transactions on Pattern Analysis and Machine Intelligence
By Vijay Mahadevan, Nuno Vasconcelos
Issue Date:January 2010
pp. 171-177
A spatiotemporal saliency algorithm based on a center-surround framework is proposed. The algorithm is inspired by biological mechanisms of motion-based perceptual grouping and extends a discriminant formulation of center-surround saliency previously propo...
 
Layered Dynamic Textures
Found in: IEEE Transactions on Pattern Analysis and Machine Intelligence
By Antoni B. Chan, Nuno Vasconcelos
Issue Date:October 2009
pp. 1862-1879
A novel video representation, the layered dynamic texture (LDT), is proposed. The LDT is a generative model, which represents a video as a collection of stochastic layers of different appearance and dynamics. Each layer is modeled as a temporal texture sam...
 
Discriminant Saliency, the Detection of Suspicious Coincidences, and Applications to Visual Recognition
Found in: IEEE Transactions on Pattern Analysis and Machine Intelligence
By Dashan Gao, Sunhyoung Han, Nuno Vasconcelos
Issue Date:June 2009
pp. 989-1005
A discriminant formulation of top-down visual saliency, intrinsically connected to the recognition problem, is proposed. The new formulation is shown to be closely related to a number of classical principles for the organization of perceptual systems, incl...
 
Tumor Targeting for Lung Cancer Radiotherapy Using Machine Learning Techniques
Found in: Machine Learning and Applications, Fourth International Conference on
By Tong Lin, Laura Cervino, Xiaoli Tang, Nuno Vasconcelos, Steve B. Jiang
Issue Date:December 2008
pp. 533-538
Accurate lung tumor targeting in real time plays a fundamental role in image-guide radiotherapy of lung cancers. Precise tumor targeting is required for both respiratory gating and tracking. Gating is considered as the current state of the art for precise ...
 
Modeling, Clustering, and Segmenting Video with Mixtures of Dynamic Textures
Found in: IEEE Transactions on Pattern Analysis and Machine Intelligence
By Antoni B. Chan, Nuno Vasconcelos
Issue Date:May 2008
pp. 909-926
A dynamic texture is a spatio-temporal generative model for video, which represents video sequences as observations from a linear dynamical system. This work studies the mixture of dynamic textures, a statistical model for an ensemble of video sequences th...
 
Supervised Learning of Semantic Classes for Image Annotation and Retrieval
Found in: IEEE Transactions on Pattern Analysis and Machine Intelligence
By Gustavo Carneiro, Antoni B. Chan, Pedro J. Moreno, Nuno Vasconcelos
Issue Date:March 2007
pp. 394-410
A probabilistic formulation for semantic image annotation and retrieval is proposed. Annotation and retrieval are posed as classification problems where each class is defined as the group of database images labeled with a common semantic label. It is shown...
 
Formulating Semantic Image Annotation as a Supervised Learning Problem
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Gustavo Carneiro, Nuno Vasconcelos
Issue Date:June 2005
pp. 163-168
We introduce a new method to automatically annotate and retrieve images using a vocabulary of image semantics. The novel contributions include a discriminant formulation of the problem, a multiple instance learning solution that enables the estimation of c...
 
Image Indexing with Mixture Hierarchies
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Nuno Vasconcelos
Issue Date:December 2001
pp. 3
We present an image indexing method based on a hierarchical description of the density of each of the image classes in a given database. The method is similar in spirit to traditional agglomerative clustering procedures but produces a complete mixture dens...
 
On the Complexity of Probabilistic Image Retrieval
Found in: Computer Vision, IEEE International Conference on
By Nuno Vasconcelos
Issue Date:July 2001
pp. 400
Probabilistic image retrieval approaches can lead to significant gains over standard retrieval techniques. However, this occurs at the cost of a significant increase in computational complexity. In fact, closed-form solutions for probabilistic retrieval ar...
 
Empirical Bayesian Motion Segmentation
Found in: IEEE Transactions on Pattern Analysis and Machine Intelligence
By Nuno Vasconcelos, Andrew Lippman
Issue Date:February 2001
pp. 217-221
<p><b>Abstract</b>—We introduce an empirical Bayesian procedure for the simultaneous segmentation of an observed motion field and estimation of the hyperparameters of a Markov random field prior. The new approach exhibits the Bayesian app...
 
A Unifying View of Image Similarity
Found in: Pattern Recognition, International Conference on
By Nuno Vasconcelos, Andrew Lippman
Issue Date:September 2000
pp. 1038
We study solutions to the problem of evaluating image similarity in the context of content-based image retrieval (CBIR). Retrieval is formulated as a classification problem, where the goal is to minimize probability of retrieval error. It is shown that thi...
 
Bayesian Relevance Feedback for Content-Based Image Retrieval
Found in: Content-Based Access of Image and Video Libraries, IEEE Workshop on
By Nuno Vasconcelos, Andrew Lippman
Issue Date:June 2000
pp. 63
We present a Bayesian learning algorithm that relies on belief propagation to integrate feedback provided by the user over a retrieval session. Bayesian retrieval leads to natural criteria for evaluating local image similarity without requiring any image s...
 
A Probabilistic Architecture for Content-Based Image Retrieval
Found in: Computer Vision and Pattern Recognition, IEEE Computer Society Conference on
By Nuno Vasconcelos, Andrew Lippman
Issue Date:June 2000
pp. 1216
The design of an effective architecture for content-based retrieval from visual libraries requires careful consideration of the interplay between feature selection, feature representation, and similarity metric. We present a solution where all the modules ...
 
Probabilistic Retrieval: New Insights and Experimental Results
Found in: Content-Based Access of Image and Video Libraries, IEEE Workshop on
By Nuno Vasconcelos, Andrew Lippman
Issue Date:June 1999
pp. 62
We present new insights on the relations between a recently introduced probabilistic formulation of the content-based retrieval problem and standard solutions. New experimental results are presented, providing evidence that probabilistic retrieval has supe...
 
A Bayesian Video Modeling Framework for Shot Segmentation and Content Characterization
Found in: Content-Based Access of Image and Video Libraries, IEEE Workshop on
By Nuno Vasconcelos, Andrew Lippman
Issue Date:June 1997
pp. 59
The segmentation of video streams into their component shots is a pre-requisite for most applications involving content-based access to video libraries. In this work, we address the segmentation problem from a probabilistic standpoint which exposes the maj...
 
On the Role of Correlation and Abstraction in Cross-Modal Multimedia Retrieval
Found in: IEEE Transactions on Pattern Analysis and Machine Intelligence
By Jose Costa Pereira,Emanuele Coviello,Gabriel Doyle,Nikhil Rasiwasia,Gert R. G. Lanckriet,Roger Levy,Nuno Vasconcelos
Issue Date:March 2014
pp. 521-535
The problem of cross-modal retrieval from multimedia repositories is considered. This problem addresses the design of retrieval systems that support queries across content modalities, for example, using an image to search for texts. A mathematical formulat...
 
Anomaly Detection and Localization in Crowded Scenes
Found in: IEEE Transactions on Pattern Analysis and Machine Intelligence
By Weixin Li,Vijay Mahadevan,Nuno Vasconcelos
Issue Date:January 2014
pp. 18-32
The detection and localization of anomalous behaviors in crowded scenes is considered, and a joint detector of temporal and spatial anomalies is proposed. The proposed detector is based on a video representation that accounts for both appearance and dynami...
 
Recognizing Activities via Bag of Words for Attribute Dynamics
Found in: 2013 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
By Weixin Li,Qian Yu,Harpreet Sawhney,Nuno Vasconcelos
Issue Date:June 2013
pp. 2587-2594
In this work, we propose a novel video representation for activity recognition that models video dynamics with attributes of activities. A video sequence is decomposed into short-term segments, which are characterized by the dynamics of their attributes. T...
 
A new approach to cross-modal multimedia retrieval
Found in: Proceedings of the international conference on Multimedia (MM '10)
By Emanuele Coviello, Gabriel Doyle, Gert R.G. Lanckriet, Jose Costa Pereira, Nikhil Rasiwasia, Nuno Vasconcelos, Roger Levy
Issue Date:October 2010
pp. 251-260
The problem of joint modeling the text and image components of multimedia documents is studied. The text component is represented as a sample from a hidden topic model, learned with latent Dirichlet allocation, and images are represented as bags of visual ...
     
Image retrieval using query by contextual example
Found in: Proceeding of the 1st ACM international conference on Multimedia information retrieval (MIR '08)
By Nikhil Rasiwasia, Nuno Vasconcelos
Issue Date:October 2008
pp. 1-1
Current image retrieval techniques have difficulties to retrieve images which exhibit distinct visual patterns but belong to the class of the query image. Previous attempts to improve generalization have shown that the introduction of semantic representati...
     
 1