The Experts below are selected from a list of 215487 Experts worldwide ranked by ideXlab platform
John Wright - One of the best experts on this subject based on the ideXlab platform.
-
efficient point to subspace query in ell 1 with application to robust object instance recognition
Siam Journal on Imaging Sciences, 2014Co-Authors: Ju Sun, Yuqian Zhang, John WrightAbstract:Motivated by vision tasks such as robust face and object recognition, we consider the following general problem: given a collection of low-dimensional linear subspaces in a high-dimensional ambient (image) space, and a query point (image), efficiently determine the nearest subspace to the query in $\ell^1$ Distance. In contrast to the naive exhaustive search which entails large-Scale linear programs, we show that the computational burden can be cut down significantly by a simple two-stage algorithm: (1) projecting the query and database subspaces into lower-dimensional space by random Cauchy matrix and solving small-Scale Distance evaluations (linear programs) in the projection space to locate the nearest candidates; (2) with few candidates upon independent repetition of (1), getting back to the high-dimensional space and performing exhaustive search. To preserve the identity of the nearest subspace with nontrivial probability, the projection dimension typically is a low-order polynomial of the subspace d...
-
efficient point to subspace query in ell 1 with application to robust object instance recognition
arXiv: Computer Vision and Pattern Recognition, 2012Co-Authors: Ju Sun, Yuqian Zhang, John WrightAbstract:Motivated by vision tasks such as robust face and object recognition, we consider the following general problem: given a collection of low-dimensional linear subspaces in a high-dimensional ambient (image) space, and a query point (image), efficiently determine the nearest subspace to the query in $\ell^1$ Distance. In contrast to the naive exhaustive search which entails large-Scale linear programs, we show that the computational burden can be cut down significantly by a simple two-stage algorithm: (1) projecting the query and data-base subspaces into lower-dimensional space by random Cauchy matrix, and solving small-Scale Distance evaluations (linear programs) in the projection space to locate candidate nearest; (2) with few candidates upon independent repetition of (1), getting back to the high-dimensional space and performing exhaustive search. To preserve the identity of the nearest subspace with nontrivial probability, the projection dimension typically is low-order polynomial of the subspace dimension multiplied by logarithm of number of the subspaces (Theorem 2.1). The reduced dimensionality and hence complexity renders the proposed algorithm particularly relevant to vision application such as robust face and object instance recognition that we investigate empirically.
Ju Sun - One of the best experts on this subject based on the ideXlab platform.
-
efficient point to subspace query in ell 1 with application to robust object instance recognition
Siam Journal on Imaging Sciences, 2014Co-Authors: Ju Sun, Yuqian Zhang, John WrightAbstract:Motivated by vision tasks such as robust face and object recognition, we consider the following general problem: given a collection of low-dimensional linear subspaces in a high-dimensional ambient (image) space, and a query point (image), efficiently determine the nearest subspace to the query in $\ell^1$ Distance. In contrast to the naive exhaustive search which entails large-Scale linear programs, we show that the computational burden can be cut down significantly by a simple two-stage algorithm: (1) projecting the query and database subspaces into lower-dimensional space by random Cauchy matrix and solving small-Scale Distance evaluations (linear programs) in the projection space to locate the nearest candidates; (2) with few candidates upon independent repetition of (1), getting back to the high-dimensional space and performing exhaustive search. To preserve the identity of the nearest subspace with nontrivial probability, the projection dimension typically is a low-order polynomial of the subspace d...
-
efficient point to subspace query in ell 1 with application to robust object instance recognition
arXiv: Computer Vision and Pattern Recognition, 2012Co-Authors: Ju Sun, Yuqian Zhang, John WrightAbstract:Motivated by vision tasks such as robust face and object recognition, we consider the following general problem: given a collection of low-dimensional linear subspaces in a high-dimensional ambient (image) space, and a query point (image), efficiently determine the nearest subspace to the query in $\ell^1$ Distance. In contrast to the naive exhaustive search which entails large-Scale linear programs, we show that the computational burden can be cut down significantly by a simple two-stage algorithm: (1) projecting the query and data-base subspaces into lower-dimensional space by random Cauchy matrix, and solving small-Scale Distance evaluations (linear programs) in the projection space to locate candidate nearest; (2) with few candidates upon independent repetition of (1), getting back to the high-dimensional space and performing exhaustive search. To preserve the identity of the nearest subspace with nontrivial probability, the projection dimension typically is low-order polynomial of the subspace dimension multiplied by logarithm of number of the subspaces (Theorem 2.1). The reduced dimensionality and hence complexity renders the proposed algorithm particularly relevant to vision application such as robust face and object instance recognition that we investigate empirically.
Jason K Blackburn - One of the best experts on this subject based on the ideXlab platform.
-
living la vida t locoh site fidelity of florida ranched and wild white tailed deer odocoileus virginianus during the epizootic hemorrhagic disease virus ehdv transmission period
Movement ecology, 2020Co-Authors: Emily T N Dinh, Allison Cauvin, Jeremy P Orange, Rebecca M Shuman, Samantha M Wisely, Jason K BlackburnAbstract:BACKGROUND: Epizootic hemorrhagic disease virus (EHDV) is a pathogen vectored by Culicoides midges that causes significant economic loss in the cervid farming industry and affects wild deer as well. Despite this, its ecology is poorly understood. Studying movement and space use by ruminant hosts during the transmission season may elucidate EHDV ecology by identifying behaviors that can increase exposure risk. Here we compared home ranges (HRs) and site fidelity metrics within HRs using the T-LoCoH R package and GPS data from collared deer. METHODS: Here, we tested whether white-tailed deer (Odocoileus virginianus) roaming within a high-fenced, private deer farm (ranched) and native deer from nearby state-managed properties (wild) exhibited differences in home range (HR) size and usage during the 2016 and 2017 EHDV seasons. We captured male and female individuals in both years and derived seasonal HRs for both sexes and both groups for each year. HRs were calculated using a time-Scale Distance approach in T-LoCoH. We then derived revisitation and duration of visit metrics and compared between years, sexes, and ranched and wild deer. RESULTS: We found that ranched deer of both sexes tended to have smaller activity spaces (95% HR) and revisited sites within their HR more often but stayed for shorter periods than wild deer. However, core area (25% HR) sizes did not significantly differ between these groups. CONCLUSIONS: The contrast in our findings between wild and ranched deer suggest that home range usage, rather than size, in addition to differences in population density, likely drive differences in disease exposure during the transmission period.
-
living la vida t locoh site fidelity of florida ranched and wild white tailed deer odocoileus virginianus during the epizootic hemorrhagic disease virus ehdv transmission period
Movement ecology, 2020Co-Authors: Emily T N Dinh, Allison Cauvin, Jeremy P Orange, Rebecca M Shuman, Samantha M Wisely, Jason K BlackburnAbstract:Epizootic hemorrhagic disease virus (EHDV) is a pathogen vectored by Culicoides midges that causes significant economic loss in the cervid farming industry and affects wild deer as well. Despite this, its ecology is poorly understood. Studying movement and space use by ruminant hosts during the transmission season may elucidate EHDV ecology by identifying behaviors that can increase exposure risk. Here we compared home ranges (HRs) and site fidelity metrics within HRs using the T-LoCoH R package and GPS data from collared deer. Here, we tested whether white-tailed deer (Odocoileus virginianus) roaming within a high-fenced, private deer farm (ranched) and native deer from nearby state-managed properties (wild) exhibited differences in home range (HR) size and usage during the 2016 and 2017 EHDV seasons. We captured male and female individuals in both years and derived seasonal HRs for both sexes and both groups for each year. HRs were calculated using a time-Scale Distance approach in T-LoCoH. We then derived revisitation and duration of visit metrics and compared between years, sexes, and ranched and wild deer. We found that ranched deer of both sexes tended to have smaller activity spaces (95% HR) and revisited sites within their HR more often but stayed for shorter periods than wild deer. However, core area (25% HR) sizes did not significantly differ between these groups. The contrast in our findings between wild and ranched deer suggest that home range usage, rather than size, in addition to differences in population density, likely drive differences in disease exposure during the transmission period.
Yuqian Zhang - One of the best experts on this subject based on the ideXlab platform.
-
efficient point to subspace query in ell 1 with application to robust object instance recognition
Siam Journal on Imaging Sciences, 2014Co-Authors: Ju Sun, Yuqian Zhang, John WrightAbstract:Motivated by vision tasks such as robust face and object recognition, we consider the following general problem: given a collection of low-dimensional linear subspaces in a high-dimensional ambient (image) space, and a query point (image), efficiently determine the nearest subspace to the query in $\ell^1$ Distance. In contrast to the naive exhaustive search which entails large-Scale linear programs, we show that the computational burden can be cut down significantly by a simple two-stage algorithm: (1) projecting the query and database subspaces into lower-dimensional space by random Cauchy matrix and solving small-Scale Distance evaluations (linear programs) in the projection space to locate the nearest candidates; (2) with few candidates upon independent repetition of (1), getting back to the high-dimensional space and performing exhaustive search. To preserve the identity of the nearest subspace with nontrivial probability, the projection dimension typically is a low-order polynomial of the subspace d...
-
efficient point to subspace query in ell 1 with application to robust object instance recognition
arXiv: Computer Vision and Pattern Recognition, 2012Co-Authors: Ju Sun, Yuqian Zhang, John WrightAbstract:Motivated by vision tasks such as robust face and object recognition, we consider the following general problem: given a collection of low-dimensional linear subspaces in a high-dimensional ambient (image) space, and a query point (image), efficiently determine the nearest subspace to the query in $\ell^1$ Distance. In contrast to the naive exhaustive search which entails large-Scale linear programs, we show that the computational burden can be cut down significantly by a simple two-stage algorithm: (1) projecting the query and data-base subspaces into lower-dimensional space by random Cauchy matrix, and solving small-Scale Distance evaluations (linear programs) in the projection space to locate candidate nearest; (2) with few candidates upon independent repetition of (1), getting back to the high-dimensional space and performing exhaustive search. To preserve the identity of the nearest subspace with nontrivial probability, the projection dimension typically is low-order polynomial of the subspace dimension multiplied by logarithm of number of the subspaces (Theorem 2.1). The reduced dimensionality and hence complexity renders the proposed algorithm particularly relevant to vision application such as robust face and object instance recognition that we investigate empirically.
Emily T N Dinh - One of the best experts on this subject based on the ideXlab platform.
-
living la vida t locoh site fidelity of florida ranched and wild white tailed deer odocoileus virginianus during the epizootic hemorrhagic disease virus ehdv transmission period
Movement ecology, 2020Co-Authors: Emily T N Dinh, Allison Cauvin, Jeremy P Orange, Rebecca M Shuman, Samantha M Wisely, Jason K BlackburnAbstract:BACKGROUND: Epizootic hemorrhagic disease virus (EHDV) is a pathogen vectored by Culicoides midges that causes significant economic loss in the cervid farming industry and affects wild deer as well. Despite this, its ecology is poorly understood. Studying movement and space use by ruminant hosts during the transmission season may elucidate EHDV ecology by identifying behaviors that can increase exposure risk. Here we compared home ranges (HRs) and site fidelity metrics within HRs using the T-LoCoH R package and GPS data from collared deer. METHODS: Here, we tested whether white-tailed deer (Odocoileus virginianus) roaming within a high-fenced, private deer farm (ranched) and native deer from nearby state-managed properties (wild) exhibited differences in home range (HR) size and usage during the 2016 and 2017 EHDV seasons. We captured male and female individuals in both years and derived seasonal HRs for both sexes and both groups for each year. HRs were calculated using a time-Scale Distance approach in T-LoCoH. We then derived revisitation and duration of visit metrics and compared between years, sexes, and ranched and wild deer. RESULTS: We found that ranched deer of both sexes tended to have smaller activity spaces (95% HR) and revisited sites within their HR more often but stayed for shorter periods than wild deer. However, core area (25% HR) sizes did not significantly differ between these groups. CONCLUSIONS: The contrast in our findings between wild and ranched deer suggest that home range usage, rather than size, in addition to differences in population density, likely drive differences in disease exposure during the transmission period.
-
living la vida t locoh site fidelity of florida ranched and wild white tailed deer odocoileus virginianus during the epizootic hemorrhagic disease virus ehdv transmission period
Movement ecology, 2020Co-Authors: Emily T N Dinh, Allison Cauvin, Jeremy P Orange, Rebecca M Shuman, Samantha M Wisely, Jason K BlackburnAbstract:Epizootic hemorrhagic disease virus (EHDV) is a pathogen vectored by Culicoides midges that causes significant economic loss in the cervid farming industry and affects wild deer as well. Despite this, its ecology is poorly understood. Studying movement and space use by ruminant hosts during the transmission season may elucidate EHDV ecology by identifying behaviors that can increase exposure risk. Here we compared home ranges (HRs) and site fidelity metrics within HRs using the T-LoCoH R package and GPS data from collared deer. Here, we tested whether white-tailed deer (Odocoileus virginianus) roaming within a high-fenced, private deer farm (ranched) and native deer from nearby state-managed properties (wild) exhibited differences in home range (HR) size and usage during the 2016 and 2017 EHDV seasons. We captured male and female individuals in both years and derived seasonal HRs for both sexes and both groups for each year. HRs were calculated using a time-Scale Distance approach in T-LoCoH. We then derived revisitation and duration of visit metrics and compared between years, sexes, and ranched and wild deer. We found that ranched deer of both sexes tended to have smaller activity spaces (95% HR) and revisited sites within their HR more often but stayed for shorter periods than wild deer. However, core area (25% HR) sizes did not significantly differ between these groups. The contrast in our findings between wild and ranched deer suggest that home range usage, rather than size, in addition to differences in population density, likely drive differences in disease exposure during the transmission period.
