JA Reggia, E. Ruppin and D. Glanzman (Eds.) Progress in Brain Research, Vol 121 1999 Elsevier Sci... more JA Reggia, E. Ruppin and D. Glanzman (Eds.) Progress in Brain Research, Vol 121 1999 Elsevier Science BV. All rights reserved, CHAPTER 10 Prosopagnosia in modular neural network models Matthew N. Dailey* and Garrison W. Cottrell Department of Computer Science and Engineering, U, C. San Diego, La Jolla, CA 92093-0] 14, USA Introduction There is strong evidence that face processing in the brain is localized.
Abstract—Autonomous target pursuit is an extremely useful technology for surveillance application... more Abstract—Autonomous target pursuit is an extremely useful technology for surveillance applications. In this paper, we derive and evaluate, in a realistic simulation, a novel tracking algorithm for vision-based pursuit. We assume a simple ground-based surveillance robot equipped with a single monocular camera. For the sensor, we propose the use of a color histogram based region tracker.
Abstract A previous experiment tested subjects' new/old judgments of previously-studied faces, di... more Abstract A previous experiment tested subjects' new/old judgments of previously-studied faces, distractors, and morphs between pairs of studied parents. We examine the extent to which models based on principal component analysis (eigenfaces) can predict human recognition of studied faces and false alarms to the distractors and morphs.
We propose and evaluate an efficient method for automatic identification of suspicious behavior i... more We propose and evaluate an efficient method for automatic identification of suspicious behavior in video surveillance data that incrementally learns scene-specific statistical models of human behavior without requiring storage of large databases of training data. The approach begins by building an initial set of models explaining the behaviors occurring in a small bootstrap dataset.
Abstract We have developed a virtual reality (VR) and an augmented reality (AR) dental training s... more Abstract We have developed a virtual reality (VR) and an augmented reality (AR) dental training simulator utilizing a haptic device. The simulators utilize volumetric force feedback computation and real time modification of the volumetric data. They include a virtual mirror to facilitate indirect vision during a simulated operation.
Numerical experiments with cooperating multiple quadratic snakes for road extraction
Higher-order active contours or snakes show much promise for the extraction of complex objects fr... more Higher-order active contours or snakes show much promise for the extraction of complex objects from noisy imagery. These models provide an elegant mathematical framework for specifying the desired properties of target objects through energy functionals that can be minimized with standard optimization techniques. However, techniques to allow quadratic snakes to change topology during segmentation have not been fully exploited.
From the user's point of view, response time is the most important quality attribute of a Web ser... more From the user's point of view, response time is the most important quality attribute of a Web service, and Web service providers aim to limit response time at minimal cost. Dynamic resource scaling in cloud computing could potentially enable specific response time guarantees, but current service-level agreements (SLAs) offered by cloud infrastructure providers do not address response time.
Shot boundary detection from videos using entropy and local descriptor
Abstract Video shot segmentation is an important step in key frame selection, video copy detectio... more Abstract Video shot segmentation is an important step in key frame selection, video copy detection, video summarization, and video indexing for retrieval. Although some types of video data, eg, live sports coverage, have abrupt shot boundaries that are easy to identify using simple heuristics, it is much more difficult to identify shot boundaries in other types such as cinematic movies.
Abstract In cloud-based Web application hosting environments, virtualization offers the potential... more Abstract In cloud-based Web application hosting environments, virtualization offers the potential to exploit dynamic resource provisioning and scaling to maintain service level agreements while minimizing resource utilization for a given workload. However, optimal proactive resource provisioning and scaling for a specific Web application require, at the least, a profile of the application's current workload and a model of the application's capacity under various resource configurations. Here we focus on multi-tier Web applications.
Abstract This paper presents the results of a series of three experiments carried out to evaluate... more Abstract This paper presents the results of a series of three experiments carried out to evaluate the speed and accuracy of target acquisition tasks using visual, auditory, or tactile feedback modalities. Ten participants conducted a series of 1-D and 2-D pointing tasks using a pen-based input device and either visual, audio or tactile (vibratory) feedback. Fitts' Law was used to analyze the results.
ABSTRACT Constructing affect-sensitive intelligent tutoring systems often requires training data ... more ABSTRACT Constructing affect-sensitive intelligent tutoring systems often requires training data in which the learner's true affective or mental states are known. We use students' self reports as ground truth data. We record students during learning then obtain self reports about their mental states during post-recording interviews. We have found a surprising level of consistency in the relationship across different students from different cultures between affective or mental states and unintentional hand gestures during learning.
A mixture of experts model exhibiting prosopagnosia
Abstract A considerable body of evidence from prosopagnosia, a deficit in face recognition dissoc... more Abstract A considerable body of evidence from prosopagnosia, a deficit in face recognition dissociable from nonface object recognition, indicates that the visual system devotes a specialized functional area to mechanisms appropriate for face processing. We present a modular neural network composed of two" expert" networks and one mediating" gate" network with the task of learning to recognize the faces of 12 individuals and classifying 36 nonface objects as members of one of three classes.
Development of a dental skills training simulator using virtual reality and haptic device
ABSTRACT–The traditional approach to dental skills training has drawbacks in terms of cost, avail... more ABSTRACT–The traditional approach to dental skills training has drawbacks in terms of cost, availability, and lack of real-world cases. We develop a virtual reality (VR) dental training system utilizing a haptic (force feedback) device. We generate three-dimensional surface model of the teeth from patient's data and a realistic model of a dental handpiece with a cylindrical cutting burr. Collision detection between virtual teeth and a tool is realized by the Axis Aligned Bounding Box (AABB) algorithm.
Abstract—We propose and evaluate a new method for automatic identification of suspicious behavior... more Abstract—We propose and evaluate a new method for automatic identification of suspicious behavior in video surveillance data. The approach works by constructing scene-specific statistical models explaining the behaviors occurring in a small bootstrap data set. It partitions the bootstrap set into clusters then assigns new observation sequences to clusters based on statistical tests of HMM log likelihood scores. Cluster-specific likelihood thresholds are learned rather than set arbitrarily.
Page 1. Cultural Differences in Facial Expression Classification Matthew N. Dailey UCSD Computer ... more Page 1. Cultural Differences in Facial Expression Classification Matthew N. Dailey UCSD Computer Science & Engineering Michael J. Lyons ATR Media Information Science Labs Miyuki Kamachi ATR Human Information Science Labs Hanae Ishi and Jiro Gyoba Tohoku University Department of Psychology Garrison W. Cottrell UCSD Computer Science and Engineering Page 2.
What are the features underlying face processing? This is a question that appears to have been as... more What are the features underlying face processing? This is a question that appears to have been asked and answered multiple times. The most recent answer appears to be that the features used in recognizing faces are holistic in nature (Biederman & Kalocsai, 1998; Farah, Wilson, Drain, & Tanaka, 1998). Generally, holistic is used to refer to a system that is sensitive to configural properties of the stimulus and that displays context-dependent interpretation of the parts (Farah et al., 1998).
Abstract Human detection and tracking in high density crowds is an unsolved problem. Standard pre... more Abstract Human detection and tracking in high density crowds is an unsolved problem. Standard preprocessing techniques such as background modeling fail when most of the scene is in motion. Because of high levels of occlusion, dense features, and shadows, object detectors tend to produce large numbers of false detections. We introduce a new method based on 3D head plane estimation that reduces these false detections while preserving high detection rates.
Abstract: We propose and evaluate a new method for automatic identification of suspicious behavio... more Abstract: We propose and evaluate a new method for automatic identification of suspicious behavior in video surveillance data. The approach works by constructing scene-specific statistical models explaining the behaviors occurring in a small bootstrap data set. It partitions the bootstrap set into clusters then assigns new observation sequences to clusters based on statistical tests of HMM log likelihood scores. Cluster-specific likelihood thresholds are learned rather than set arbitrarily.
Adaptive resource provisioning for read intensive multi-tier applications in the cloud
Abstract A Service-Level Agreement (SLA) provides surety for specific quality attributes to the c... more Abstract A Service-Level Agreement (SLA) provides surety for specific quality attributes to the consumers of services. However, current SLAs offered by cloud infrastructure providers do not address response time, which, from the user's point of view, is the most important quality attribute for Web applications.
Exudates are among the preliminary signs of diabetic retinopathy, a major cause of vision loss in... more Exudates are among the preliminary signs of diabetic retinopathy, a major cause of vision loss in diabetic patients. Early detection of exudates could improve patients' chances to avoid blindness. In this paper, we present a series of experiments on feature selection and exudates classification using naive Bayes and support vector machine (SVM) classifiers.
JA Reggia, E. Ruppin and D. Glanzman (Eds.) Progress in Brain Research, Vol 121 1999 Elsevier Sci... more JA Reggia, E. Ruppin and D. Glanzman (Eds.) Progress in Brain Research, Vol 121 1999 Elsevier Science BV. All rights reserved, CHAPTER 10 Prosopagnosia in modular neural network models Matthew N. Dailey* and Garrison W. Cottrell Department of Computer Science and Engineering, U, C. San Diego, La Jolla, CA 92093-0] 14, USA Introduction There is strong evidence that face processing in the brain is localized.
Abstract—Autonomous target pursuit is an extremely useful technology for surveillance application... more Abstract—Autonomous target pursuit is an extremely useful technology for surveillance applications. In this paper, we derive and evaluate, in a realistic simulation, a novel tracking algorithm for vision-based pursuit. We assume a simple ground-based surveillance robot equipped with a single monocular camera. For the sensor, we propose the use of a color histogram based region tracker.
Abstract A previous experiment tested subjects' new/old judgments of previously-studied faces, di... more Abstract A previous experiment tested subjects' new/old judgments of previously-studied faces, distractors, and morphs between pairs of studied parents. We examine the extent to which models based on principal component analysis (eigenfaces) can predict human recognition of studied faces and false alarms to the distractors and morphs.
We propose and evaluate an efficient method for automatic identification of suspicious behavior i... more We propose and evaluate an efficient method for automatic identification of suspicious behavior in video surveillance data that incrementally learns scene-specific statistical models of human behavior without requiring storage of large databases of training data. The approach begins by building an initial set of models explaining the behaviors occurring in a small bootstrap dataset.
Abstract We have developed a virtual reality (VR) and an augmented reality (AR) dental training s... more Abstract We have developed a virtual reality (VR) and an augmented reality (AR) dental training simulator utilizing a haptic device. The simulators utilize volumetric force feedback computation and real time modification of the volumetric data. They include a virtual mirror to facilitate indirect vision during a simulated operation.
Numerical experiments with cooperating multiple quadratic snakes for road extraction
Higher-order active contours or snakes show much promise for the extraction of complex objects fr... more Higher-order active contours or snakes show much promise for the extraction of complex objects from noisy imagery. These models provide an elegant mathematical framework for specifying the desired properties of target objects through energy functionals that can be minimized with standard optimization techniques. However, techniques to allow quadratic snakes to change topology during segmentation have not been fully exploited.
From the user's point of view, response time is the most important quality attribute of a Web ser... more From the user's point of view, response time is the most important quality attribute of a Web service, and Web service providers aim to limit response time at minimal cost. Dynamic resource scaling in cloud computing could potentially enable specific response time guarantees, but current service-level agreements (SLAs) offered by cloud infrastructure providers do not address response time.
Shot boundary detection from videos using entropy and local descriptor
Abstract Video shot segmentation is an important step in key frame selection, video copy detectio... more Abstract Video shot segmentation is an important step in key frame selection, video copy detection, video summarization, and video indexing for retrieval. Although some types of video data, eg, live sports coverage, have abrupt shot boundaries that are easy to identify using simple heuristics, it is much more difficult to identify shot boundaries in other types such as cinematic movies.
Abstract In cloud-based Web application hosting environments, virtualization offers the potential... more Abstract In cloud-based Web application hosting environments, virtualization offers the potential to exploit dynamic resource provisioning and scaling to maintain service level agreements while minimizing resource utilization for a given workload. However, optimal proactive resource provisioning and scaling for a specific Web application require, at the least, a profile of the application's current workload and a model of the application's capacity under various resource configurations. Here we focus on multi-tier Web applications.
Abstract This paper presents the results of a series of three experiments carried out to evaluate... more Abstract This paper presents the results of a series of three experiments carried out to evaluate the speed and accuracy of target acquisition tasks using visual, auditory, or tactile feedback modalities. Ten participants conducted a series of 1-D and 2-D pointing tasks using a pen-based input device and either visual, audio or tactile (vibratory) feedback. Fitts' Law was used to analyze the results.
ABSTRACT Constructing affect-sensitive intelligent tutoring systems often requires training data ... more ABSTRACT Constructing affect-sensitive intelligent tutoring systems often requires training data in which the learner's true affective or mental states are known. We use students' self reports as ground truth data. We record students during learning then obtain self reports about their mental states during post-recording interviews. We have found a surprising level of consistency in the relationship across different students from different cultures between affective or mental states and unintentional hand gestures during learning.
A mixture of experts model exhibiting prosopagnosia
Abstract A considerable body of evidence from prosopagnosia, a deficit in face recognition dissoc... more Abstract A considerable body of evidence from prosopagnosia, a deficit in face recognition dissociable from nonface object recognition, indicates that the visual system devotes a specialized functional area to mechanisms appropriate for face processing. We present a modular neural network composed of two" expert" networks and one mediating" gate" network with the task of learning to recognize the faces of 12 individuals and classifying 36 nonface objects as members of one of three classes.
Development of a dental skills training simulator using virtual reality and haptic device
ABSTRACT–The traditional approach to dental skills training has drawbacks in terms of cost, avail... more ABSTRACT–The traditional approach to dental skills training has drawbacks in terms of cost, availability, and lack of real-world cases. We develop a virtual reality (VR) dental training system utilizing a haptic (force feedback) device. We generate three-dimensional surface model of the teeth from patient's data and a realistic model of a dental handpiece with a cylindrical cutting burr. Collision detection between virtual teeth and a tool is realized by the Axis Aligned Bounding Box (AABB) algorithm.
Abstract—We propose and evaluate a new method for automatic identification of suspicious behavior... more Abstract—We propose and evaluate a new method for automatic identification of suspicious behavior in video surveillance data. The approach works by constructing scene-specific statistical models explaining the behaviors occurring in a small bootstrap data set. It partitions the bootstrap set into clusters then assigns new observation sequences to clusters based on statistical tests of HMM log likelihood scores. Cluster-specific likelihood thresholds are learned rather than set arbitrarily.
Page 1. Cultural Differences in Facial Expression Classification Matthew N. Dailey UCSD Computer ... more Page 1. Cultural Differences in Facial Expression Classification Matthew N. Dailey UCSD Computer Science & Engineering Michael J. Lyons ATR Media Information Science Labs Miyuki Kamachi ATR Human Information Science Labs Hanae Ishi and Jiro Gyoba Tohoku University Department of Psychology Garrison W. Cottrell UCSD Computer Science and Engineering Page 2.
What are the features underlying face processing? This is a question that appears to have been as... more What are the features underlying face processing? This is a question that appears to have been asked and answered multiple times. The most recent answer appears to be that the features used in recognizing faces are holistic in nature (Biederman & Kalocsai, 1998; Farah, Wilson, Drain, & Tanaka, 1998). Generally, holistic is used to refer to a system that is sensitive to configural properties of the stimulus and that displays context-dependent interpretation of the parts (Farah et al., 1998).
Abstract Human detection and tracking in high density crowds is an unsolved problem. Standard pre... more Abstract Human detection and tracking in high density crowds is an unsolved problem. Standard preprocessing techniques such as background modeling fail when most of the scene is in motion. Because of high levels of occlusion, dense features, and shadows, object detectors tend to produce large numbers of false detections. We introduce a new method based on 3D head plane estimation that reduces these false detections while preserving high detection rates.
Abstract: We propose and evaluate a new method for automatic identification of suspicious behavio... more Abstract: We propose and evaluate a new method for automatic identification of suspicious behavior in video surveillance data. The approach works by constructing scene-specific statistical models explaining the behaviors occurring in a small bootstrap data set. It partitions the bootstrap set into clusters then assigns new observation sequences to clusters based on statistical tests of HMM log likelihood scores. Cluster-specific likelihood thresholds are learned rather than set arbitrarily.
Adaptive resource provisioning for read intensive multi-tier applications in the cloud
Abstract A Service-Level Agreement (SLA) provides surety for specific quality attributes to the c... more Abstract A Service-Level Agreement (SLA) provides surety for specific quality attributes to the consumers of services. However, current SLAs offered by cloud infrastructure providers do not address response time, which, from the user's point of view, is the most important quality attribute for Web applications.
Exudates are among the preliminary signs of diabetic retinopathy, a major cause of vision loss in... more Exudates are among the preliminary signs of diabetic retinopathy, a major cause of vision loss in diabetic patients. Early detection of exudates could improve patients' chances to avoid blindness. In this paper, we present a series of experiments on feature selection and exudates classification using naive Bayes and support vector machine (SVM) classifiers.
Uploads
Papers by Matthew Dailey