In this paper, we present statistical models to predict the difficulty of recognizing musical keys from polyphonic audio signals. Automatic audio key finding has been studied for many years, and various approaches have been proposed and reported. Reports of these methods' performance are usually based on the proposers' own data sets. Without details on the data set, i.e., how challenging the data set is, directly comparing the effectiveness of these methods is not meaningful or even possible. Thus, in this study we focus on predicting the difficulty level of key recognition as perceived by human experts. Given an audio recording, represented as the extracted acoustic features, we apply multiple linear regression and proportional odds model to predict the difficulty level of the recording, annotated by experts as an integer on a 5-point Likert scale. We use four metrics to evaluate our prediction results: root mean square error, Pearson correlation coefficient, exact accuracy, and adjacent accuracy. We also examine the difference between experts' annotations and discuss their consistency.
{"title":"Predicting Key Recognition Difficulty in Polyphonic Audio","authors":"C. Chuan, Aleksey Charapko","doi":"10.1109/ISM.2013.82","DOIUrl":"https://doi.org/10.1109/ISM.2013.82","url":null,"abstract":"In this paper, we present statistical models to predict the difficulty of recognizing musical keys from polyphonic audio signals. Automatic audio key finding has been studied for many years, and various approaches have been proposed and reported. Reports of these methods' performance are usually based on the proposers' own data sets. Without details on the data set, i.e., how challenging the data set is, directly comparing the effectiveness of these methods is not meaningful or even possible. Thus, in this study we focus on predicting the difficulty level of key recognition as perceived by human experts. Given an audio recording, represented as the extracted acoustic features, we apply multiple linear regression and proportional odds model to predict the difficulty level of the recording, annotated by experts as an integer on a 5-point Likert scale. We use four metrics to evaluate our prediction results: root mean square error, Pearson correlation coefficient, exact accuracy, and adjacent accuracy. We also examine the difference between experts' annotations and discuss their consistency.","PeriodicalId":6311,"journal":{"name":"2013 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB)","volume":"3 1","pages":"421-426"},"PeriodicalIF":0.0,"publicationDate":"2013-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83620748","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}
Visual matching for tracking and recognition, for example in video indexing, often uses image features measured at multiple resolutions. As a tracked object moves away from the camera, appearing progressively smaller, the higher resolutions consecutively become unavailable for matching, causing step changes in the similarity or “match score” of the tracked object. If several candidate matches (hypotheses) are maintained for a tracked region, this effect causes a bias toward larger region hypotheses that match at one extra resolution relative to even slightly smaller hypotheses. The effect is subtle and appears intermittent because it occurs only around a specific discrete set of object sizes. We describe the problem and the class of visual matching methods that it affects, and propose a solution. We present experimental results from a real video indexing system to illustrate both the problem and the effectiveness of the proposed solution.
{"title":"Resolution Control for Size Bias Elimination in Multi-resolution Visual Matching","authors":"S. Clippingdale","doi":"10.1109/ISM.2013.87","DOIUrl":"https://doi.org/10.1109/ISM.2013.87","url":null,"abstract":"Visual matching for tracking and recognition, for example in video indexing, often uses image features measured at multiple resolutions. As a tracked object moves away from the camera, appearing progressively smaller, the higher resolutions consecutively become unavailable for matching, causing step changes in the similarity or “match score” of the tracked object. If several candidate matches (hypotheses) are maintained for a tracked region, this effect causes a bias toward larger region hypotheses that match at one extra resolution relative to even slightly smaller hypotheses. The effect is subtle and appears intermittent because it occurs only around a specific discrete set of object sizes. We describe the problem and the class of visual matching methods that it affects, and propose a solution. We present experimental results from a real video indexing system to illustrate both the problem and the effectiveness of the proposed solution.","PeriodicalId":6311,"journal":{"name":"2013 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB)","volume":"76 1","pages":"451-456"},"PeriodicalIF":0.0,"publicationDate":"2013-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83857076","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}
Tamara Seybold, Christian Keimel, Marion Knopp, W. Stechele
The development and tuning of denoising algorithms is usually based on readily processed test images that are artificially degraded with additive white Gaussian noise (AWGN). While AWGN allows us to easily generate test data in a repeatable manner, it does not reflect the noise characteristics in a real digital camera. Realistic camera noise is signal-dependent and spatially correlated due to the demosaicking step required to obtain full-color images. Hence, the noise characteristic is fundamentally different from AWGN. Using such unrealistic data to test, optimize and compare denoising algorithms may lead to incorrect parameter tuning or sub optimal choices in research on denoising algorithms. In this paper, we therefore propose an approach to evaluate denoising algorithms with respect to realistic camera noise: we describe a new camera noise model that includes the full processing chain of a single sensor camera. We determine the visual quality of noisy and denoised test sequences using a subjective test with 18 participants. We show that the noise characteristics have a significant effect on visual quality. Quality metrics, which are required to compare denoising results, are applied, and we evaluate the performance of 10 full-reference metrics and one no-reference metric with our realistic test data. We conclude that a more realistic noise model should be used in future research to improve the quality estimation of digital images and videos and to improve the research on denoising algorithms.
{"title":"Towards an Evaluation of Denoising Algorithms with Respect to Realistic Camera Noise","authors":"Tamara Seybold, Christian Keimel, Marion Knopp, W. Stechele","doi":"10.1109/ISM.2013.39","DOIUrl":"https://doi.org/10.1109/ISM.2013.39","url":null,"abstract":"The development and tuning of denoising algorithms is usually based on readily processed test images that are artificially degraded with additive white Gaussian noise (AWGN). While AWGN allows us to easily generate test data in a repeatable manner, it does not reflect the noise characteristics in a real digital camera. Realistic camera noise is signal-dependent and spatially correlated due to the demosaicking step required to obtain full-color images. Hence, the noise characteristic is fundamentally different from AWGN. Using such unrealistic data to test, optimize and compare denoising algorithms may lead to incorrect parameter tuning or sub optimal choices in research on denoising algorithms. In this paper, we therefore propose an approach to evaluate denoising algorithms with respect to realistic camera noise: we describe a new camera noise model that includes the full processing chain of a single sensor camera. We determine the visual quality of noisy and denoised test sequences using a subjective test with 18 participants. We show that the noise characteristics have a significant effect on visual quality. Quality metrics, which are required to compare denoising results, are applied, and we evaluate the performance of 10 full-reference metrics and one no-reference metric with our realistic test data. We conclude that a more realistic noise model should be used in future research to improve the quality estimation of digital images and videos and to improve the research on denoising algorithms.","PeriodicalId":6311,"journal":{"name":"2013 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB)","volume":"29 1","pages":"203-210"},"PeriodicalIF":0.0,"publicationDate":"2013-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83582587","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}
Despite the recent success of extensive co-segmentation studies, they still suffer from limitations in accommodating multiple-foreground, large-scale, high-variability image set, as well as their underlying capability for parallel implementation. To improve, this paper proposes a bi-harmonic distance governed flexible method for the robust coherent segmentation of the overlapping/similar contents co-existing in image group, which is independent of supervised learning and any other user-specified prior. The central idea is the novel integration of bi-harmonic distance metric design and multi-level deformable graph generation for multi-level clustering, which gives rise to a host of unique advantages: accommodating multiple-foreground images, respecting both local structures and global semantics of images, being more robust and accurate, and being convenient for parallel acceleration. Critical pipeline of our method involves intrinsic content-coherent measuring, super-pixel assisted bottom-up clustering, and multi-level deformable graph clustering based cross-image optimization. We conduct extensive experiments on the iCoseg benchmark and Oxford flower datasets, and make comprehensive evaluations to demonstrate the superiority of our method via comparison with state-of-the-art methods collected in the MSRC database.
{"title":"Unsupervised Co-segmentation of Complex Image Set via Bi-harmonic Distance Governed Multi-level Deformable Graph Clustering","authors":"Jizhou Ma, Shuai Li, A. Hao, Hong Qin","doi":"10.1109/ISM.2013.16","DOIUrl":"https://doi.org/10.1109/ISM.2013.16","url":null,"abstract":"Despite the recent success of extensive co-segmentation studies, they still suffer from limitations in accommodating multiple-foreground, large-scale, high-variability image set, as well as their underlying capability for parallel implementation. To improve, this paper proposes a bi-harmonic distance governed flexible method for the robust coherent segmentation of the overlapping/similar contents co-existing in image group, which is independent of supervised learning and any other user-specified prior. The central idea is the novel integration of bi-harmonic distance metric design and multi-level deformable graph generation for multi-level clustering, which gives rise to a host of unique advantages: accommodating multiple-foreground images, respecting both local structures and global semantics of images, being more robust and accurate, and being convenient for parallel acceleration. Critical pipeline of our method involves intrinsic content-coherent measuring, super-pixel assisted bottom-up clustering, and multi-level deformable graph clustering based cross-image optimization. We conduct extensive experiments on the iCoseg benchmark and Oxford flower datasets, and make comprehensive evaluations to demonstrate the superiority of our method via comparison with state-of-the-art methods collected in the MSRC database.","PeriodicalId":6311,"journal":{"name":"2013 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB)","volume":"50 1","pages":"38-45"},"PeriodicalIF":0.0,"publicationDate":"2013-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83976155","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}
Jan Vanvinkenroye, Christoph Grüninger, C. Heine, T. Richter
We implemented a survey with one learning group using the web-based tools and a control group working with a traditional setup based on editor and compiler. In a recent publication, we described the design and implementation of a web-based programming lab (ViPLab) targeted at undergraduate Engineering and Mathematics courses. This work provides a quantitative analysis of the user feedback, experience and learning success. The survey shows that web-based installations are as efficient as classical tools, while Windows users prefer the web-based chain over the editor/compiler installation on Linux. This justifies the use of web-based installations in programming beginner courses, if the learning target focuses on programming and not a particular tool chain.
{"title":"A Quantitative Analysis of a Virtual Programming Lab","authors":"Jan Vanvinkenroye, Christoph Grüninger, C. Heine, T. Richter","doi":"10.1109/ISM.2013.88","DOIUrl":"https://doi.org/10.1109/ISM.2013.88","url":null,"abstract":"We implemented a survey with one learning group using the web-based tools and a control group working with a traditional setup based on editor and compiler. In a recent publication, we described the design and implementation of a web-based programming lab (ViPLab) targeted at undergraduate Engineering and Mathematics courses. This work provides a quantitative analysis of the user feedback, experience and learning success. The survey shows that web-based installations are as efficient as classical tools, while Windows users prefer the web-based chain over the editor/compiler installation on Linux. This justifies the use of web-based installations in programming beginner courses, if the learning target focuses on programming and not a particular tool chain.","PeriodicalId":6311,"journal":{"name":"2013 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB)","volume":"61 1","pages":"457-461"},"PeriodicalIF":0.0,"publicationDate":"2013-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80731895","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}
The quality of synthetic speech offered by pitch and duration modifications via Time Domain Pitch Synchronous Overlap Add method (TD-PSOLA) relies on an accurate positioning of pitch marks. In this paper, we propose a new pitch marking technique of voiced regions based on the fundamental signal of the speech waveform. By using the valleys of the fundamental signal, we locate a set of precise intervals where the exact instants of pitch marks are expected to be found. The fundamental signal is composed only from the fundamental frequency (pitch) of speech. It is represented by a specific signal named "mean based signal" (MBS). The optimal pitch marks are found by extracting the set of global peak instants within the obtained intervals. To improve the performance of the proposed technique, we have proposed a post processing stage which allows us to correct the erroneous pitch marks that may occur due to some synchronization problems. The proposed technique is evaluated on CMU ACRTIC database by using objective and subjective measures. The experiments demonstrate that the proposed technique allows pitch and duration modifications via TD-PSOLA with high quality.
{"title":"Pitch Marking Using the Fundamental Signal for Speech Modifications via TDPSOLA","authors":"F. Ykhlef, L. Bendaouia","doi":"10.1109/ISM.2013.28","DOIUrl":"https://doi.org/10.1109/ISM.2013.28","url":null,"abstract":"The quality of synthetic speech offered by pitch and duration modifications via Time Domain Pitch Synchronous Overlap Add method (TD-PSOLA) relies on an accurate positioning of pitch marks. In this paper, we propose a new pitch marking technique of voiced regions based on the fundamental signal of the speech waveform. By using the valleys of the fundamental signal, we locate a set of precise intervals where the exact instants of pitch marks are expected to be found. The fundamental signal is composed only from the fundamental frequency (pitch) of speech. It is represented by a specific signal named \"mean based signal\" (MBS). The optimal pitch marks are found by extracting the set of global peak instants within the obtained intervals. To improve the performance of the proposed technique, we have proposed a post processing stage which allows us to correct the erroneous pitch marks that may occur due to some synchronization problems. The proposed technique is evaluated on CMU ACRTIC database by using objective and subjective measures. The experiments demonstrate that the proposed technique allows pitch and duration modifications via TD-PSOLA with high quality.","PeriodicalId":6311,"journal":{"name":"2013 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB)","volume":"57 1","pages":"118-124"},"PeriodicalIF":0.0,"publicationDate":"2013-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80227193","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}
Automated detection of moving objects is an essential task for any intelligent transportation system. However, conventional motion detection techniques often suffer from the loss of moving objects due to bit-rate variation in video streams transmitted via wireless video communication systems. To achieve motion detection that is both reliable and accurate in video streams of variable bit-rate, this paper proposes a novel motion detection approach which is based on grey relational analysis, and which integrates a multi-quality background generation module and a moving object detection module. As our experimental results demonstrate, the proposed approach attained superior motion detection performance compared to other state-of-the-art techniques based on qualitative and quantitative evaluations. Quantitative evaluations produced F1 and Similarity accuracy scores for the proposed approach that were up to 59.96% and 55.42% higher than those of the other compared techniques, respectively.
{"title":"Accurate Detection of Moving Objects in Traffic Video Streams over Limited Bandwidth Networks","authors":"Bo-Hao Chen, Shih-Chia Huang","doi":"10.1109/ISM.2013.20","DOIUrl":"https://doi.org/10.1109/ISM.2013.20","url":null,"abstract":"Automated detection of moving objects is an essential task for any intelligent transportation system. However, conventional motion detection techniques often suffer from the loss of moving objects due to bit-rate variation in video streams transmitted via wireless video communication systems. To achieve motion detection that is both reliable and accurate in video streams of variable bit-rate, this paper proposes a novel motion detection approach which is based on grey relational analysis, and which integrates a multi-quality background generation module and a moving object detection module. As our experimental results demonstrate, the proposed approach attained superior motion detection performance compared to other state-of-the-art techniques based on qualitative and quantitative evaluations. Quantitative evaluations produced F1 and Similarity accuracy scores for the proposed approach that were up to 59.96% and 55.42% higher than those of the other compared techniques, respectively.","PeriodicalId":6311,"journal":{"name":"2013 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB)","volume":"13 1","pages":"69-75"},"PeriodicalIF":0.0,"publicationDate":"2013-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85439868","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}
Digital Object Repository in the Digital Object Architecture stores a large number of audio/video media files. Lack of metadata in audio/video media files limits the media discovery service in Digital Object Architecture from searching those media files. In this paper we designed a system that uses name spotting module to extract the names, stores the extracted names with audio/video media files, simulates the media discovery service and reports the findings related to the improvement in searching the media file.
{"title":"An Improvement in Media Discovery Service Using Name Spotting","authors":"Manish Goswami, Lan Yang","doi":"10.1109/ISM.2013.83","DOIUrl":"https://doi.org/10.1109/ISM.2013.83","url":null,"abstract":"Digital Object Repository in the Digital Object Architecture stores a large number of audio/video media files. Lack of metadata in audio/video media files limits the media discovery service in Digital Object Architecture from searching those media files. In this paper we designed a system that uses name spotting module to extract the names, stores the extracted names with audio/video media files, simulates the media discovery service and reports the findings related to the improvement in searching the media file.","PeriodicalId":6311,"journal":{"name":"2013 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB)","volume":"10 1","pages":"427-432"},"PeriodicalIF":0.0,"publicationDate":"2013-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90252177","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 investigated the potential of the real-time implementation in single image super resolution using edge directed unsharp masking sharpening (EDUMS) method. To achieve efficient real-time implementation with unsharp masking sharpening, the resolution enhancement process needed only simply filtering operations without iterations. Also, with edge directed information as the prior of the unsharp masking sharpening method, the jaggy artifact was efficiently suppressed. Clear edge structures and vivid details of high resolution images with minimum artifacts were presented by the proposed method.
{"title":"Efficient Super Resolution Using Edge Directed Unsharp Masking Sharpening Method","authors":"Kuo-Shiuan Peng, F. Lin, Yi-Pai Huang, H. Shieh","doi":"10.1109/ISM.2013.100","DOIUrl":"https://doi.org/10.1109/ISM.2013.100","url":null,"abstract":"This paper investigated the potential of the real-time implementation in single image super resolution using edge directed unsharp masking sharpening (EDUMS) method. To achieve efficient real-time implementation with unsharp masking sharpening, the resolution enhancement process needed only simply filtering operations without iterations. Also, with edge directed information as the prior of the unsharp masking sharpening method, the jaggy artifact was efficiently suppressed. Clear edge structures and vivid details of high resolution images with minimum artifacts were presented by the proposed method.","PeriodicalId":6311,"journal":{"name":"2013 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB)","volume":"65 1","pages":"508-509"},"PeriodicalIF":0.0,"publicationDate":"2013-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73364173","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}
In this paper a novel approach to the mixed noise removal in color images is proposed. The described method is a generalization of the Non-Local Means algorithm, where the pixels in the filtering window are ordered and only the most centrally located pixels in the filtering window are considered and used to calculate the weights needed for the averaging operation. The comparison with the existing state-of-the-art denoising schemes in terms of image restoration quality measures shows, that the new approach yields significantly better results in suppressing mixed noise in color digital images.
{"title":"Trimmed Non-local Means Technique for Mixed Noise Removal in Color Images","authors":"Krystian Radlak, B. Smolka","doi":"10.1109/ISM.2013.78","DOIUrl":"https://doi.org/10.1109/ISM.2013.78","url":null,"abstract":"In this paper a novel approach to the mixed noise removal in color images is proposed. The described method is a generalization of the Non-Local Means algorithm, where the pixels in the filtering window are ordered and only the most centrally located pixels in the filtering window are considered and used to calculate the weights needed for the averaging operation. The comparison with the existing state-of-the-art denoising schemes in terms of image restoration quality measures shows, that the new approach yields significantly better results in suppressing mixed noise in color digital images.","PeriodicalId":6311,"journal":{"name":"2013 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB)","volume":"13 1","pages":"405-406"},"PeriodicalIF":0.0,"publicationDate":"2013-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76826743","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：