Human neurocristopathies include a number of syndromes, tumors, and dysmorphologies of neural crest （NC） stem cell derivatives. In recent years, many white spotting genes have been associated with hypopigmentary dis...Human neurocristopathies include a number of syndromes, tumors, and dysmorphologies of neural crest （NC） stem cell derivatives. In recent years, many white spotting genes have been associated with hypopigmentary disorders and deafness in neurocristopathies resulting from NC stem cell-derived melanocyte deficiency during development. These include PAX3, SOX10, MITF, SNAI2, EDNRB, EDN3, KIT, and KITL. Recent studies have revealed surprising new insights into a central role of MITF in the complex network of interacting genes in melanocyte development. In this perspective, we provide an overview of some of the current findings and explore complex functional roles of these genes during NC stem cell-derived melanocyte development.展开更多
End-to-end text spotting is a vital computer vision task that aims to integrate scene text detection and recognition into a unified framework.Typical methods heavily rely on region-of-interest(Rol)operations to extrac...End-to-end text spotting is a vital computer vision task that aims to integrate scene text detection and recognition into a unified framework.Typical methods heavily rely on region-of-interest(Rol)operations to extract local features and complex post-processing steps to produce final predictions.To address these limitations,we propose TextFormer,a query-based end-to-end text spotter with a transformer architecture.Specifically,using query embedding per text instance,TextFormer builds upon an image encoder and a text decoder to learn a joint semantic understanding for multitask modeling.It allows for mutual training and optimization of classification,segmentation and recognition branches,resulting in deeper feature sharing without sacrificing flexibility or simplicity.Additionally,we design an adaptive global aggregation(AGG)module to transfer global features into sequential features for reading arbitrarilyshaped texts,which overcomes the suboptimization problem of Rol operations.Furthermore,potential corpus information is utilized from weak annotations to full labels through mixed supervision,further improving text detection and end-to-end text spotting results.Extensive experiments on various bilingual(i.e.,English and Chinese)benchmarks demonstrate the superiority of our method.Especially on the TDA-ReCTS dataset,TextFormer surpasses the state-of-the-art method in terms of 1-NED by 13.2%.展开更多
Keyword Spotting (KWS) or Keyword Recognition is to recognize a set of given keywords in continuous,spontaneous speech.It is a very active branch of Speech Recognition.It differs from Continuous Speech Recognition (CS...Keyword Spotting (KWS) or Keyword Recognition is to recognize a set of given keywords in continuous,spontaneous speech.It is a very active branch of Speech Recognition.It differs from Continuous Speech Recognition (CSR) in terms of research focus,implementation techniques and application background.So they are often regarded as two different issues.This paper gives an overview of basic techniques and advances in recent years,including how to combine the traits of Mandarin.展开更多
in this paper a novel technique adopted in HarkMan is introduced. HarkMan is a keyword-spotter designed to automatically spot the given words of a vocabulary-independent task in unconstrained Chinese telephone speech....in this paper a novel technique adopted in HarkMan is introduced. HarkMan is a keyword-spotter designed to automatically spot the given words of a vocabulary-independent task in unconstrained Chinese telephone speech. The speak- ing manner and the number of keywords are not limited. This paper focuses on the novel technique which addresses acoustic modeling, keyword spotting network, search strategies, robustness, and rejection. The underlying technologies used in HarkMan given in this paper are useful not only for keyword spotting but also for continuous speech recognition. The system has achieved a figure-of-merit value over 90%.展开更多
N-ethyl-N-nitrosourea （ENU） mutagenesis has led to the elucidation of several regulator genes for melanocyte and skin development. Here we characterized a mutant from ENU mutagenesis with similar phenotype as that o...N-ethyl-N-nitrosourea （ENU） mutagenesis has led to the elucidation of several regulator genes for melanocyte and skin development. Here we characterized a mutant from ENU mutagenesis with similar phenotype as that of Splotch mutant, including exencephaly, spina bifida and abnormal limbs in homozygotes as well as white belly spotting and occasionally loop-tail in heterozygotes. This novel mutant was named as SpxG. Through genome-wide linkage analysis in backcross progenies with microsatellite markers, the SpxG was confined to a region between DIMIT415 and DIMIT7 on chromosome 1, where notable Pax3 gene was located. Direct sequencing revealed that SpxG carried a nucleotide A894G missense transition in exon 6 of Pax3 gene that resulted in Asn to Asp substitution at amino acid 269 within the highly-conserved homeodomain （HD） DNA recognition module, which was the first point mutation found in this domain in mice. This N269D mutation impaired the transactivation capacity of Pax3 protein, but exerted no effect on Pax3 protein translation. The characterization of the new mutation expanded our understanding the transactivation and DNA-binding structure of Pax3 protein.展开更多
Chinese calligraphy is a very special style of handwriting and direct character recognition is very difficult. Content-based keyword spotting is more feasible than recognition-based retrieval for calligraphy document....Chinese calligraphy is a very special style of handwriting and direct character recognition is very difficult. Content-based keyword spotting is more feasible than recognition-based retrieval for calligraphy document. In this paper,we propose a novel Elastic Histogram of Oriented Gradient( EHOG) descriptor for calligraphy word spotting. The presented feature is a modification of Histogram of Oriented Gradient( HOG), widely used in human detection. In our approach,the input word image is partitioned into non-uniform rectangular cells according to the calligraphy character pixel intensity,and then in each cell a histogram of orientation is accumulated dynamically. Moreover,we adopt Derivative Dynamic Time Warping( DDTW) for image feature matching,which achieves good performance in gesture recognition. Experiments demonstrate a very significant improvement when comparing our proposed feature with previously developed ones,and also show DDTW produces superior alignments between two calligraphy character feature series than DTW.展开更多
We develop a new technique to measure the exact upper diameters of trees that is comparatively simple and inexpensive. We can measure the diameters of entire tree trunks efficiently and with high precision. The system...We develop a new technique to measure the exact upper diameters of trees that is comparatively simple and inexpensive. We can measure the diameters of entire tree trunks efficiently and with high precision. The system uses a digital camera with a ~15 - 30× telephoto lens to take a photograph that can be used for measuring the diameter of the upper part of a comparatively slender tree trunk. Since this method requires a measuring distance and the height of the target point in the image, a range finder capable of measuring angles was combined with the main digiscoping system. A range finder sensor uses a laser and makes a 360 degree angle of observation possible. The diameter of a target position of the objective tree can be obtained by measuring the digital image using image editing software and calculations from spreadsheet software. We focus on the Japanese cedar species in the southwestern part of Japan. Photographic measurements were obtained prior to thinning. The estimates that we obtained largely agree with the true measurements of all trees. With regard to the estimated accuracy of all measured trees, the maximum error ratio was 7.0% (1.45 cm), with a ~2% - 4% error for most of the estimated results. Although the absolute value of the estimation error was 1.87 cm (5.3%) at the maximum (9.87 m in height and 35.5 cm in diameter), an estimation accuracy of <1 cm was reproduced in almost all measurements except for the extreme hypertrophy portion by butt swelling.展开更多
A new scheme is presented to detect a large number ofKeywordsin voice controlled switchboard tasks. The new scheme is based on two stages. In the first stage, N best syllable candidates with their corresponding acous...A new scheme is presented to detect a large number ofKeywordsin voice controlled switchboard tasks. The new scheme is based on two stages. In the first stage, N best syllable candidates with their corresponding acoustic scores are generated by an acoustic recognizer. In the second stage, a semantic model based parser is applied to determine the optimum keywords by searching through the lattice of N best candidates. The experimental results show that when the spoken input deviates from the predefined syntactic constraints, the parser can also demonstrate high performance. For comparison purposes, the most common way to incorporate the syntactic knowledge of the task directly into the acoustic recognizer in the form of a finite state network is also investigated. Furthermore, to address the sparse data problems, out of domain data in the form of newspaper text are used to obtain a more robust combined semantic model. The experiments show that the combined semantic model can improve the keywords detection rate from 90.07% to 92.91% when 80 ungrammatical sentences which do not conform to the task grammar are used as testing material.展开更多
In 2006, Methodist Le Bonheur Healthcare (MLH) created the Congregational Health Network (CHN, TM pending) which works closely with clergy in the most under-served zip codes of the city to improve access to care and o...In 2006, Methodist Le Bonheur Healthcare (MLH) created the Congregational Health Network (CHN, TM pending) which works closely with clergy in the most under-served zip codes of the city to improve access to care and overall health status of the population. To best coordinate CHN resources around high-utilization and address the largest health needs in the community, MLH applied hot spotting and geographic information system (GIS) spatial analysis techniques. These techniques were coupled with the community health needs assessment process at MLH and qualitative, participatory research findings captured in collaboration with church and other community partners. The methodology, which we call “participatory hot spotting,” is based upon the Camden Model, which leverages hot spotting to assess and prioritize community need in the provision of charity care, but adds a participatory, qualitative layer. In this study, spatial analysis was employed to evaluate hospital-based inpatient and outpatient utilization and define costs of charity care for the health system by area of residence. Ten zip codes accounted for 56% of total system charity care costs. Among these, the largest zip code, as defined by a percentage of total charity costs, contributed 18% of the inpatient utilization and 17% of the cost. Further, this zip code (38109) contributed 69% of the inpatient and 76% of the outpatient charity care volume and accounted for 75% of inpatient and 76% of outpatient charity care costs for the system. These findings were combined with grassroots intelligence that enabled a partnership with clergy and community members and Cigna Healthcare to better coordinate care in a place-based population health management strategy. Presentations of the analytics have subsequently been made to HHS and the CDC, referred to by some as the “Memphis Model”.展开更多
Facial micro-expressions are short and imperceptible expressions that involuntarily reveal the true emotions that a person may be attempting to suppress,hide,disguise,or conceal.Such expressions can reflect a person&#...Facial micro-expressions are short and imperceptible expressions that involuntarily reveal the true emotions that a person may be attempting to suppress,hide,disguise,or conceal.Such expressions can reflect a person's real emotions and have a wide range of application in public safety and clinical diagnosis.The analysis of facial micro-expressions in video sequences through computer vision is still relatively recent.In this research,a comprehensive review on the topic of spotting and recognition used in micro expression analysis databases and methods,is conducted,and advanced technologies in this area are summarized.In addition,we discuss challenges that remain unresolved alongside future work to be completed in the field of micro-expression analysis.展开更多
基金We would like to thank Dr HeinzAmheiter （NIH/NINDS） for generously contributing to the images, and Dr Laura Baxter and Dr Yingzi Yang （NIH/NHGRI） for thoughtful comments on the manuscript. We also acknowledge the support by the National Natural Science Foundation of China （30771149） and the Intramural Research Program of the National Human Genome Research Institute, National Institutes of Health.
文摘Human neurocristopathies include a number of syndromes, tumors, and dysmorphologies of neural crest （NC） stem cell derivatives. In recent years, many white spotting genes have been associated with hypopigmentary disorders and deafness in neurocristopathies resulting from NC stem cell-derived melanocyte deficiency during development. These include PAX3, SOX10, MITF, SNAI2, EDNRB, EDN3, KIT, and KITL. Recent studies have revealed surprising new insights into a central role of MITF in the complex network of interacting genes in melanocyte development. In this perspective, we provide an overview of some of the current findings and explore complex functional roles of these genes during NC stem cell-derived melanocyte development.
基金supported by the National Natural Science Foundation of China(No.61902027).
文摘End-to-end text spotting is a vital computer vision task that aims to integrate scene text detection and recognition into a unified framework.Typical methods heavily rely on region-of-interest(Rol)operations to extract local features and complex post-processing steps to produce final predictions.To address these limitations,we propose TextFormer,a query-based end-to-end text spotter with a transformer architecture.Specifically,using query embedding per text instance,TextFormer builds upon an image encoder and a text decoder to learn a joint semantic understanding for multitask modeling.It allows for mutual training and optimization of classification,segmentation and recognition branches,resulting in deeper feature sharing without sacrificing flexibility or simplicity.Additionally,we design an adaptive global aggregation(AGG)module to transfer global features into sequential features for reading arbitrarilyshaped texts,which overcomes the suboptimization problem of Rol operations.Furthermore,potential corpus information is utilized from weak annotations to full labels through mixed supervision,further improving text detection and end-to-end text spotting results.Extensive experiments on various bilingual(i.e.,English and Chinese)benchmarks demonstrate the superiority of our method.Especially on the TDA-ReCTS dataset,TextFormer surpasses the state-of-the-art method in terms of 1-NED by 13.2%.
文摘Keyword Spotting (KWS) or Keyword Recognition is to recognize a set of given keywords in continuous,spontaneous speech.It is a very active branch of Speech Recognition.It differs from Continuous Speech Recognition (CSR) in terms of research focus,implementation techniques and application background.So they are often regarded as two different issues.This paper gives an overview of basic techniques and advances in recent years,including how to combine the traits of Mandarin.
文摘in this paper a novel technique adopted in HarkMan is introduced. HarkMan is a keyword-spotter designed to automatically spot the given words of a vocabulary-independent task in unconstrained Chinese telephone speech. The speak- ing manner and the number of keywords are not limited. This paper focuses on the novel technique which addresses acoustic modeling, keyword spotting network, search strategies, robustness, and rejection. The underlying technologies used in HarkMan given in this paper are useful not only for keyword spotting but also for continuous speech recognition. The system has achieved a figure-of-merit value over 90%.
基金supported by the National Basic Research Program of China(No.2007CB947301)the National Natural Science Foundation of China(No.30800613)Pujiang Talent(No.08PJ1407200)
文摘N-ethyl-N-nitrosourea （ENU） mutagenesis has led to the elucidation of several regulator genes for melanocyte and skin development. Here we characterized a mutant from ENU mutagenesis with similar phenotype as that of Splotch mutant, including exencephaly, spina bifida and abnormal limbs in homozygotes as well as white belly spotting and occasionally loop-tail in heterozygotes. This novel mutant was named as SpxG. Through genome-wide linkage analysis in backcross progenies with microsatellite markers, the SpxG was confined to a region between DIMIT415 and DIMIT7 on chromosome 1, where notable Pax3 gene was located. Direct sequencing revealed that SpxG carried a nucleotide A894G missense transition in exon 6 of Pax3 gene that resulted in Asn to Asp substitution at amino acid 269 within the highly-conserved homeodomain （HD） DNA recognition module, which was the first point mutation found in this domain in mice. This N269D mutation impaired the transactivation capacity of Pax3 protein, but exerted no effect on Pax3 protein translation. The characterization of the new mutation expanded our understanding the transactivation and DNA-binding structure of Pax3 protein.
基金Sponsored by the National Natural Science Foundation of China(Grant No.61173086)Shandong Excellent Young Scientist Award Fund(Grant No.BS2011DX002)Shandong Province Science and Technology Development Planning(Grant No.2012GSF12105)
文摘Chinese calligraphy is a very special style of handwriting and direct character recognition is very difficult. Content-based keyword spotting is more feasible than recognition-based retrieval for calligraphy document. In this paper,we propose a novel Elastic Histogram of Oriented Gradient( EHOG) descriptor for calligraphy word spotting. The presented feature is a modification of Histogram of Oriented Gradient( HOG), widely used in human detection. In our approach,the input word image is partitioned into non-uniform rectangular cells according to the calligraphy character pixel intensity,and then in each cell a histogram of orientation is accumulated dynamically. Moreover,we adopt Derivative Dynamic Time Warping( DDTW) for image feature matching,which achieves good performance in gesture recognition. Experiments demonstrate a very significant improvement when comparing our proposed feature with previously developed ones,and also show DDTW produces superior alignments between two calligraphy character feature series than DTW.
文摘We develop a new technique to measure the exact upper diameters of trees that is comparatively simple and inexpensive. We can measure the diameters of entire tree trunks efficiently and with high precision. The system uses a digital camera with a ~15 - 30× telephoto lens to take a photograph that can be used for measuring the diameter of the upper part of a comparatively slender tree trunk. Since this method requires a measuring distance and the height of the target point in the image, a range finder capable of measuring angles was combined with the main digiscoping system. A range finder sensor uses a laser and makes a 360 degree angle of observation possible. The diameter of a target position of the objective tree can be obtained by measuring the digital image using image editing software and calculations from spreadsheet software. We focus on the Japanese cedar species in the southwestern part of Japan. Photographic measurements were obtained prior to thinning. The estimates that we obtained largely agree with the true measurements of all trees. With regard to the estimated accuracy of all measured trees, the maximum error ratio was 7.0% (1.45 cm), with a ~2% - 4% error for most of the estimated results. Although the absolute value of the estimation error was 1.87 cm (5.3%) at the maximum (9.87 m in height and 35.5 cm in diameter), an estimation accuracy of <1 cm was reproduced in almost all measurements except for the extreme hypertrophy portion by butt swelling.
基金the State High-Tech Developments Planof China !( No. 863 -3 0 6-0 2 -1) Chinese211Engineering Project!( No.9610 3 -2 )
文摘A new scheme is presented to detect a large number ofKeywordsin voice controlled switchboard tasks. The new scheme is based on two stages. In the first stage, N best syllable candidates with their corresponding acoustic scores are generated by an acoustic recognizer. In the second stage, a semantic model based parser is applied to determine the optimum keywords by searching through the lattice of N best candidates. The experimental results show that when the spoken input deviates from the predefined syntactic constraints, the parser can also demonstrate high performance. For comparison purposes, the most common way to incorporate the syntactic knowledge of the task directly into the acoustic recognizer in the form of a finite state network is also investigated. Furthermore, to address the sparse data problems, out of domain data in the form of newspaper text are used to obtain a more robust combined semantic model. The experiments show that the combined semantic model can improve the keywords detection rate from 90.07% to 92.91% when 80 ungrammatical sentences which do not conform to the task grammar are used as testing material.
文摘In 2006, Methodist Le Bonheur Healthcare (MLH) created the Congregational Health Network (CHN, TM pending) which works closely with clergy in the most under-served zip codes of the city to improve access to care and overall health status of the population. To best coordinate CHN resources around high-utilization and address the largest health needs in the community, MLH applied hot spotting and geographic information system (GIS) spatial analysis techniques. These techniques were coupled with the community health needs assessment process at MLH and qualitative, participatory research findings captured in collaboration with church and other community partners. The methodology, which we call “participatory hot spotting,” is based upon the Camden Model, which leverages hot spotting to assess and prioritize community need in the provision of charity care, but adds a participatory, qualitative layer. In this study, spatial analysis was employed to evaluate hospital-based inpatient and outpatient utilization and define costs of charity care for the health system by area of residence. Ten zip codes accounted for 56% of total system charity care costs. Among these, the largest zip code, as defined by a percentage of total charity costs, contributed 18% of the inpatient utilization and 17% of the cost. Further, this zip code (38109) contributed 69% of the inpatient and 76% of the outpatient charity care volume and accounted for 75% of inpatient and 76% of outpatient charity care costs for the system. These findings were combined with grassroots intelligence that enabled a partnership with clergy and community members and Cigna Healthcare to better coordinate care in a place-based population health management strategy. Presentations of the analytics have subsequently been made to HHS and the CDC, referred to by some as the “Memphis Model”.
文摘Facial micro-expressions are short and imperceptible expressions that involuntarily reveal the true emotions that a person may be attempting to suppress,hide,disguise,or conceal.Such expressions can reflect a person's real emotions and have a wide range of application in public safety and clinical diagnosis.The analysis of facial micro-expressions in video sequences through computer vision is still relatively recent.In this research,a comprehensive review on the topic of spotting and recognition used in micro expression analysis databases and methods,is conducted,and advanced technologies in this area are summarized.In addition,we discuss challenges that remain unresolved alongside future work to be completed in the field of micro-expression analysis.
