Pub Date : 2009-06-20DOI: 10.1109/CVPRW.2009.5204052
Yu Sun, B. Bhanu, Shiv Bhanu
This paper presents a fully automated symmetry-integrated brain injury detection method for magnetic resonance imaging (MRI) sequences. One of the limitations of current injury detection methods often involves a large amount of training data or a prior model that is only applicable to a limited domain of brain slices, with low computational efficiency and robustness. Our proposed approach can detect injuries from a wide variety of brain images since it makes use of symmetry as a dominant feature, and does not rely on any prior models and training phases. The approach consists of the following steps: (a) symmetry integrated segmentation of brain slices based on symmetry affinity matrix, (b) computation of kurtosis and skewness of symmetry affinity matrix to find potential asymmetric regions, (c) clustering of the pixels in symmetry affinity matrix using a 3D relaxation algorithm, (d) fusion of the results of (b) and (c) to obtain refined asymmetric regions, (e) Gaussian mixture model for unsupervised classification of potential asymmetric regions as the set of regions corresponding to brain injuries. Experimental results are carried out to demonstrate the efficacy of the approach.
{"title":"Automatic symmetry-integrated brain injury detection in MRI sequences","authors":"Yu Sun, B. Bhanu, Shiv Bhanu","doi":"10.1109/CVPRW.2009.5204052","DOIUrl":"https://doi.org/10.1109/CVPRW.2009.5204052","url":null,"abstract":"This paper presents a fully automated symmetry-integrated brain injury detection method for magnetic resonance imaging (MRI) sequences. One of the limitations of current injury detection methods often involves a large amount of training data or a prior model that is only applicable to a limited domain of brain slices, with low computational efficiency and robustness. Our proposed approach can detect injuries from a wide variety of brain images since it makes use of symmetry as a dominant feature, and does not rely on any prior models and training phases. The approach consists of the following steps: (a) symmetry integrated segmentation of brain slices based on symmetry affinity matrix, (b) computation of kurtosis and skewness of symmetry affinity matrix to find potential asymmetric regions, (c) clustering of the pixels in symmetry affinity matrix using a 3D relaxation algorithm, (d) fusion of the results of (b) and (c) to obtain refined asymmetric regions, (e) Gaussian mixture model for unsupervised classification of potential asymmetric regions as the set of regions corresponding to brain injuries. Experimental results are carried out to demonstrate the efficacy of the approach.","PeriodicalId":431981,"journal":{"name":"2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132318944","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-06-20DOI: 10.1109/CVPRW.2009.5204357
Hongwen Kang, Alexei A. Efros, M. Hebert, T. Kanade
In this paper, we propose a data driven approach to first-person vision. We propose a novel image matching algorithm, named Re-Search, that is designed to cope with self-repetitive structures and confusing patterns in the indoor environment. This algorithm uses state-of-art image search techniques, and it matches a query image with a two-pass strategy. In the first pass, a conventional image search algorithm is used to search for a small number of images that are most similar to the query image. In the second pass, the retrieval results from the first step are used to discover features that are more distinctive in the local context. We demonstrate and evaluate the Re-Search algorithm in the context of indoor localization, with the illustration of potential applications in object pop-out and data-driven zoom-in.
{"title":"Image matching in large scale indoor environment","authors":"Hongwen Kang, Alexei A. Efros, M. Hebert, T. Kanade","doi":"10.1109/CVPRW.2009.5204357","DOIUrl":"https://doi.org/10.1109/CVPRW.2009.5204357","url":null,"abstract":"In this paper, we propose a data driven approach to first-person vision. We propose a novel image matching algorithm, named Re-Search, that is designed to cope with self-repetitive structures and confusing patterns in the indoor environment. This algorithm uses state-of-art image search techniques, and it matches a query image with a two-pass strategy. In the first pass, a conventional image search algorithm is used to search for a small number of images that are most similar to the query image. In the second pass, the retrieval results from the first step are used to discover features that are more distinctive in the local context. We demonstrate and evaluate the Re-Search algorithm in the context of indoor localization, with the illustration of potential applications in object pop-out and data-driven zoom-in.","PeriodicalId":431981,"journal":{"name":"2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"113965266","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-06-20DOI: 10.1109/CVPRW.2009.5204275
Lei Zhang, Q. Ji
Global shape prior knowledge is a special kind of semantic information that can be incorporated into an image segmentation process to handle the difficulties caused by such problems as occlusion, cluttering, noise, and/or low contrast boundaries. In this work, we propose a global shape prior representation and incorporate it into a level set based image segmentation framework. This global shape prior can effectively help remove the cluttered elongate structures and island-like artifacts from the evolving contours. We apply this global shape prior to segmentation of three sequences of electron tomography membrane images. The segmentation results are evaluated both quantitatively and qualitatively by visual inspection. Accurate segmentation results are achieved in the testing sequences, which demonstrates the capability of the proposed global shape prior representation.
{"title":"A level set-based global shape prior and its application to image segmentation","authors":"Lei Zhang, Q. Ji","doi":"10.1109/CVPRW.2009.5204275","DOIUrl":"https://doi.org/10.1109/CVPRW.2009.5204275","url":null,"abstract":"Global shape prior knowledge is a special kind of semantic information that can be incorporated into an image segmentation process to handle the difficulties caused by such problems as occlusion, cluttering, noise, and/or low contrast boundaries. In this work, we propose a global shape prior representation and incorporate it into a level set based image segmentation framework. This global shape prior can effectively help remove the cluttered elongate structures and island-like artifacts from the evolving contours. We apply this global shape prior to segmentation of three sequences of electron tomography membrane images. The segmentation results are evaluated both quantitatively and qualitatively by visual inspection. Accurate segmentation results are achieved in the testing sequences, which demonstrates the capability of the proposed global shape prior representation.","PeriodicalId":431981,"journal":{"name":"2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123845411","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-06-20DOI: 10.1109/CVPRW.2009.5204311
A. Gyaourova, A. Ross
In biometric identification systems, the identity associated with the input data is determined by comparing it against every entry in the database. This exhaustive matching process increases the response time of the system and, potentially, the rate of erroneous identification. A method that narrows the list of potential identities will allow the input data to be matched against a smaller number of identities. We describe a method for indexing large-scale multimodal biometric databases based on the generation of an index code for each enrolled identity. In the proposed method, the input biometric data is first matched against a small set of reference images. The set of ensuing match scores is used as an index code. The index codes of multiple modalities are then integrated using three different fusion techniques in order to further improve the indexing performance. Experiments on a chimeric face and fingerprint bimodal database indicate a 76% reduction in the search space at 100% hit rate. These results suggest that indexing has the potential to substantially improve the response time of multimodal biometric systems without compromising the accuracy of identification.
{"title":"A coding scheme for indexing multimodal biometric databases","authors":"A. Gyaourova, A. Ross","doi":"10.1109/CVPRW.2009.5204311","DOIUrl":"https://doi.org/10.1109/CVPRW.2009.5204311","url":null,"abstract":"In biometric identification systems, the identity associated with the input data is determined by comparing it against every entry in the database. This exhaustive matching process increases the response time of the system and, potentially, the rate of erroneous identification. A method that narrows the list of potential identities will allow the input data to be matched against a smaller number of identities. We describe a method for indexing large-scale multimodal biometric databases based on the generation of an index code for each enrolled identity. In the proposed method, the input biometric data is first matched against a small set of reference images. The set of ensuing match scores is used as an index code. The index codes of multiple modalities are then integrated using three different fusion techniques in order to further improve the indexing performance. Experiments on a chimeric face and fingerprint bimodal database indicate a 76% reduction in the search space at 100% hit rate. These results suggest that indexing has the potential to substantially improve the response time of multimodal biometric systems without compromising the accuracy of identification.","PeriodicalId":431981,"journal":{"name":"2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops","volume":"187 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123920830","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-06-20DOI: 10.1109/CVPRW.2009.5204310
Jinyu Zuo, N. Schmid
In the field of iris-based recognition, evaluation of quality of images has a number of important applications. These include image acquisition, enhancement, and data fusion. Iris image quality metrics designed for these applications are used as figures of merit to quantify degradations or improvements in iris images due to various image processing operations. This paper elaborates on the factors and introduces new global and local factors that can be used to evaluate iris video and image quality. The main contributions of the paper are as follows. (1) A fast global quality evaluation procedure for selecting the best frames from a video or an image sequence is introduced. (2) A number of new local quality measures for the iris biometrics are introduced. The performance of the individual quality measures is carefully analyzed. Since performance of iris recognition systems is evaluated in terms of the distributions of matching scores and recognition probability of error, from a good iris image quality metric it is also expected that its performance is linked to the recognition performance of the biometric recognition system.
{"title":"Global and local quality measures for NIR iris video","authors":"Jinyu Zuo, N. Schmid","doi":"10.1109/CVPRW.2009.5204310","DOIUrl":"https://doi.org/10.1109/CVPRW.2009.5204310","url":null,"abstract":"In the field of iris-based recognition, evaluation of quality of images has a number of important applications. These include image acquisition, enhancement, and data fusion. Iris image quality metrics designed for these applications are used as figures of merit to quantify degradations or improvements in iris images due to various image processing operations. This paper elaborates on the factors and introduces new global and local factors that can be used to evaluate iris video and image quality. The main contributions of the paper are as follows. (1) A fast global quality evaluation procedure for selecting the best frames from a video or an image sequence is introduced. (2) A number of new local quality measures for the iris biometrics are introduced. The performance of the individual quality measures is carefully analyzed. Since performance of iris recognition systems is evaluated in terms of the distributions of matching scores and recognition probability of error, from a good iris image quality metric it is also expected that its performance is linked to the recognition performance of the biometric recognition system.","PeriodicalId":431981,"journal":{"name":"2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops","volume":"355 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115931296","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-06-20DOI: 10.1109/CVPRW.2009.5204360
Xiaofeng Ren, Matthai Philipose
Recognizing objects being manipulated in hands can provide essential information about a person's activities and have far-reaching impacts on the application of vision in everyday life. The egocentric viewpoint from a wearable camera has unique advantages in recognizing handled objects, such as having a close view and seeing objects in their natural positions. We collect a comprehensive dataset and analyze the feasibilities and challenges of the egocentric recognition of handled objects. We use a lapel-worn camera and record uncompressed video streams as human subjects manipulate objects in daily activities. We use 42 day-to-day objects that vary in size, shape, color and textureness. 10 video sequences are shot for each object under different illuminations and backgrounds. We use this dataset and a SIFT-based recognition system to analyze and quantitatively characterize the main challenges in egocentric object recognition, such as motion blur and hand occlusion, along with its unique constraints, such as hand color, location prior and temporal consistency. SIFT-based recognition has an average recognition rate of 12%, and reaches 20% through enforcing temporal consistency. We use simulations to estimate the upper bound for SIFT-based recognition at 64%, the loss of accuracy due to background clutter at 20%, and that of hand occlusion at 13%. Our quantitative evaluations show that the egocentric recognition of handled objects is a challenging but feasible problem with many unique characteristics and many opportunities for future research.
{"title":"Egocentric recognition of handled objects: Benchmark and analysis","authors":"Xiaofeng Ren, Matthai Philipose","doi":"10.1109/CVPRW.2009.5204360","DOIUrl":"https://doi.org/10.1109/CVPRW.2009.5204360","url":null,"abstract":"Recognizing objects being manipulated in hands can provide essential information about a person's activities and have far-reaching impacts on the application of vision in everyday life. The egocentric viewpoint from a wearable camera has unique advantages in recognizing handled objects, such as having a close view and seeing objects in their natural positions. We collect a comprehensive dataset and analyze the feasibilities and challenges of the egocentric recognition of handled objects. We use a lapel-worn camera and record uncompressed video streams as human subjects manipulate objects in daily activities. We use 42 day-to-day objects that vary in size, shape, color and textureness. 10 video sequences are shot for each object under different illuminations and backgrounds. We use this dataset and a SIFT-based recognition system to analyze and quantitatively characterize the main challenges in egocentric object recognition, such as motion blur and hand occlusion, along with its unique constraints, such as hand color, location prior and temporal consistency. SIFT-based recognition has an average recognition rate of 12%, and reaches 20% through enforcing temporal consistency. We use simulations to estimate the upper bound for SIFT-based recognition at 64%, the loss of accuracy due to background clutter at 20%, and that of hand occlusion at 13%. Our quantitative evaluations show that the egocentric recognition of handled objects is a challenging but feasible problem with many unique characteristics and many opportunities for future research.","PeriodicalId":431981,"journal":{"name":"2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops","volume":"80 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122960861","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-06-20DOI: 10.1109/CVPRW.2009.5204051
Shaoting Zhang, Jinghao Zhou, Xiaoxu Wang, Sukmoon Chang, Dimitris N. Metaxas, George J. Pappas, F. Delis, N. Volkow, Gene-Jack Wang, P. Thanos, C. Kambhamettu
3D functional segmentation of brain images is important in understating the relationships between anatomy and mental diseases in brains. Volumetric analysis of various brain structures such as the cerebellum plays a critical role in studying the structural changes in brain regions as a function of development, trauma, or neurodegeneration. Although various segmentation methods in clinical studies have been proposed, many of them require a priori knowledge about the locations of the structures of interest, which prevents the fully automatic segmentation. Besides, the topological changes of structures are difficult to detect. In this paper, we present a novel method for detecting and locating the brain structures of interest that can be used for the fully automatic 3D functional segmentation of rodent brain MR images. The presented method is based on active shape model (ASM), Metamorph models and variational techniques. It focuses on detecting the topological changes of brain structures based on a novel volume ratio criteria. The mean successful rate of the topological change detection shows 86.6% accuracy compared to the expert identified ground truth.
{"title":"3D segmentation of rodent brains using deformable models and variational methods","authors":"Shaoting Zhang, Jinghao Zhou, Xiaoxu Wang, Sukmoon Chang, Dimitris N. Metaxas, George J. Pappas, F. Delis, N. Volkow, Gene-Jack Wang, P. Thanos, C. Kambhamettu","doi":"10.1109/CVPRW.2009.5204051","DOIUrl":"https://doi.org/10.1109/CVPRW.2009.5204051","url":null,"abstract":"3D functional segmentation of brain images is important in understating the relationships between anatomy and mental diseases in brains. Volumetric analysis of various brain structures such as the cerebellum plays a critical role in studying the structural changes in brain regions as a function of development, trauma, or neurodegeneration. Although various segmentation methods in clinical studies have been proposed, many of them require a priori knowledge about the locations of the structures of interest, which prevents the fully automatic segmentation. Besides, the topological changes of structures are difficult to detect. In this paper, we present a novel method for detecting and locating the brain structures of interest that can be used for the fully automatic 3D functional segmentation of rodent brain MR images. The presented method is based on active shape model (ASM), Metamorph models and variational techniques. It focuses on detecting the topological changes of brain structures based on a novel volume ratio criteria. The mean successful rate of the topological change detection shows 86.6% accuracy compared to the expert identified ground truth.","PeriodicalId":431981,"journal":{"name":"2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126440900","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-06-20DOI: 10.1109/CVPRW.2009.5204346
S. Madan, Kristin J. Dana, G. O. Cula
In quantitative dermatology, high resolution sensors provide images that capture fine scale features like pores, birthmarks, and moles. Breathing and minute movements result in misregistration of micro level features. Many computer vision methods for dermatology such as change detection, appearance capture, and multi sensor fusion require high accuracy point-wise registration of micro level features. However, most computer vision algorithms are based on macro level features such as eyes, nose, and lips, and aren't suitable for registering micro level features. In this paper, we develop a practical robust algorithm to align face regions using skin texture with mostly indistinct micro level features. In computer vision, these regions would typically be considered featureless regions. Our method approximates the face surface as a collection of quasi-planar skin patches and uses quasiconvex optimization and the L∞ norm for estimation of spatially varying homographies. We have assembled a unique dataset of high resolution dermatology images comprised of over 100 human subjects. The image pairs vary in imaging modality (crossed, parallel and no polarization) and are misregistered due to the natural non-rigid human movement between image capture. This method of polarization based image capture is commonly used in dermatology to image surface and subsurface structure. Using this dataset, we show high quality alignment of “featureless” regions and demonstrate that the algorithm works robustly over a large set of subjects with different skin texture appearance, not just a few test images.
{"title":"Quasiconvex alignment of multimodal skin images for quantitative dermatology","authors":"S. Madan, Kristin J. Dana, G. O. Cula","doi":"10.1109/CVPRW.2009.5204346","DOIUrl":"https://doi.org/10.1109/CVPRW.2009.5204346","url":null,"abstract":"In quantitative dermatology, high resolution sensors provide images that capture fine scale features like pores, birthmarks, and moles. Breathing and minute movements result in misregistration of micro level features. Many computer vision methods for dermatology such as change detection, appearance capture, and multi sensor fusion require high accuracy point-wise registration of micro level features. However, most computer vision algorithms are based on macro level features such as eyes, nose, and lips, and aren't suitable for registering micro level features. In this paper, we develop a practical robust algorithm to align face regions using skin texture with mostly indistinct micro level features. In computer vision, these regions would typically be considered featureless regions. Our method approximates the face surface as a collection of quasi-planar skin patches and uses quasiconvex optimization and the L∞ norm for estimation of spatially varying homographies. We have assembled a unique dataset of high resolution dermatology images comprised of over 100 human subjects. The image pairs vary in imaging modality (crossed, parallel and no polarization) and are misregistered due to the natural non-rigid human movement between image capture. This method of polarization based image capture is commonly used in dermatology to image surface and subsurface structure. Using this dataset, we show high quality alignment of “featureless” regions and demonstrate that the algorithm works robustly over a large set of subjects with different skin texture appearance, not just a few test images.","PeriodicalId":431981,"journal":{"name":"2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114686771","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2009-06-20DOI: 10.1109/cvpr.2009.5204316
P. Barnum, S. Narasimhan, T. Kanade
Various non-traditional media, such as water drops, mist, and fire, have been used to create vibrant two and three dimensional displays. Usually such displays require a great deal of design and engineering. In this work, we show a computer vision based approach to easily calibrate and learn the properties of a three-dimensional water drop display, using a few pieces of off-the-shelf hardware. Our setup consists of a camera, projector, laser plane, and water drop generator. Based on the geometric calibration between the hardware, a user can “paint” the drops from the point of view of the camera, causing the projector to illuminate them with the correct color at the correct time. We first demonstrate an algorithm for the case where no drop occludes another from the point of view of either camera or projector. If there is no occlusion, the system can be trained once, and the projector plays a precomputed movie. We then show our work toward a display with real rain. In real time, our system tracks and predicts the future location of hundreds of drops per second, then projects rays to hit or miss each drop.
{"title":"A projector-camera system for creating a display with water drops","authors":"P. Barnum, S. Narasimhan, T. Kanade","doi":"10.1109/cvpr.2009.5204316","DOIUrl":"https://doi.org/10.1109/cvpr.2009.5204316","url":null,"abstract":"Various non-traditional media, such as water drops, mist, and fire, have been used to create vibrant two and three dimensional displays. Usually such displays require a great deal of design and engineering. In this work, we show a computer vision based approach to easily calibrate and learn the properties of a three-dimensional water drop display, using a few pieces of off-the-shelf hardware. Our setup consists of a camera, projector, laser plane, and water drop generator. Based on the geometric calibration between the hardware, a user can “paint” the drops from the point of view of the camera, causing the projector to illuminate them with the correct color at the correct time. We first demonstrate an algorithm for the case where no drop occludes another from the point of view of either camera or projector. If there is no occlusion, the system can be trained once, and the projector plays a precomputed movie. We then show our work toward a display with real rain. In real time, our system tracks and predicts the future location of hundreds of drops per second, then projects rays to hit or miss each drop.","PeriodicalId":431981,"journal":{"name":"2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops","volume":"66 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121864976","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper reports a method for acquiring the prior probability of human existence by using past human trajectories and the color of an image. The priors play important roles in human detection as well as in scene understanding. The proposed method is based on the assumption that a person can exist again in an area where he/she existed in the past. In order to acquire the priors efficiently, a high prior probability is assigned to an area having the same color as past human trajectories. We use a particle filter for representing the prior probability. Therefore, we can represent a complex prior probability using only a few parameters. Through experiments, we confirmed that our proposed method can acquire the prior probability efficiently and it can realize highly accurate human detection using the obtained prior probability.
{"title":"Efficient acquisition of human existence priors from motion trajectories","authors":"H. Habe, Hidehito Nakagawa, M. Kidode","doi":"10.2197/ipsjtcva.2.145","DOIUrl":"https://doi.org/10.2197/ipsjtcva.2.145","url":null,"abstract":"This paper reports a method for acquiring the prior probability of human existence by using past human trajectories and the color of an image. The priors play important roles in human detection as well as in scene understanding. The proposed method is based on the assumption that a person can exist again in an area where he/she existed in the past. In order to acquire the priors efficiently, a high prior probability is assigned to an area having the same color as past human trajectories. We use a particle filter for representing the prior probability. Therefore, we can represent a complex prior probability using only a few parameters. Through experiments, we confirmed that our proposed method can acquire the prior probability efficiently and it can realize highly accurate human detection using the obtained prior probability.","PeriodicalId":431981,"journal":{"name":"2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133757466","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：