Zobrazeno 1 - 10
of 18
pro vyhledávání: '"Breton Minnehan"'
Publikováno v:
IEEE Open Journal of Signal Processing, Vol 1, Pp 90-98 (2020)
Principal component analysis is one of the most commonly used methods for dimensionality reduction in signal processing. However, the most commonly used PCA formulation is based on the L2-norm, which can be highly influenced by outlier data. In recen
Externí odkaz:
https://doaj.org/article/44ab3b0e04044d3f9bfe11c91a18ab30
Autor:
Navya Nagananda, Abu Md Niamul Taufique, Raaga Madappa, Chowdhury Sadman Jahan, Breton Minnehan, Todd Rovito, Andreas Savakis
Publikováno v:
Sensors, Vol 21, Iss 23, p 8070 (2021)
Deep learning grew in importance in recent years due to its versatility and excellent performance on supervised classification tasks. A core assumption for such supervised approaches is that the training and testing data are drawn from the same under
Externí odkaz:
https://doaj.org/article/b29666a701c84c0fa8eb796715e60213
Publikováno v:
Sensors, Vol 20, Iss 2, p 547 (2020)
In recent years, deep learning-based visual object trackers have achieved state-of-the-art performance on several visual object tracking benchmarks. However, most tracking benchmarks are focused on ground level videos, whereas aerial tracking present
Externí odkaz:
https://doaj.org/article/cc35c7fa7d0f48b7b4a15d03ea75dade
Publikováno v:
Journal of Applied Remote Sensing. 16
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::3ca903c7d403d608949145e070433b26
https://doi.org/10.1117/1.jrs.16.036501
https://doi.org/10.1117/1.jrs.16.036501
Autor:
Andreas Savakis, Navya Nagananda, Chowdhury Sadman Jahan, Raaga Madappa, Breton Minnehan, Todd V. Rovito, Abu Md Niamul Taufique
Publikováno v:
Sensors, Vol 21, Iss 8070, p 8070 (2021)
Sensors (Basel, Switzerland)
Sensors; Volume 21; Issue 23; Pages: 8070
Sensors (Basel, Switzerland)
Sensors; Volume 21; Issue 23; Pages: 8070
Deep learning grew in importance in recent years due to its versatility and excellent performance on supervised classification tasks. A core assumption for such supervised approaches is that the training and testing data are drawn from the same under
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::9fda150e7d22cf2999f38ff7baabacd4
https://www.mdpi.com/1424-8220/21/23/8070
https://www.mdpi.com/1424-8220/21/23/8070
Autor:
Andreas Savakis, Breton Minnehan
Publikováno v:
Machine Vision and Applications. 30:473-485
We propose a novel deep learning domain adaptation method that performs transductive learning from the source domain to the target domain based on cluster matching between the source and target features. The proposed method combines Adaptive Batch No
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::739c7113d077888adb0cc9d31f864a33
https://doi.org/10.1007/s00138-019-01003-1
https://doi.org/10.1007/s00138-019-01003-1
Autor:
Breton Minnehan, Andreas Savakis
Publikováno v:
IEEE Signal Processing Letters. 26:242-246
In this letter, we propose a fast Grassmann manifold optimization method for $L_1$ -norm based principal component analysis (GM- $L_1$ -PCA). Our approach is a two-step iterative cost-minimization and manifold retraction technique that efficiently fi
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::721970a4d926c1a7b2a823430d4d8c1b
https://doi.org/10.1109/lsp.2018.2886742
https://doi.org/10.1109/lsp.2018.2886742
Publikováno v:
Geospatial Informatics IX.
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::7dda7a4d1ea4078f468c2302aec23c26
https://doi.org/10.1117/12.2518823
https://doi.org/10.1117/12.2518823
Autor:
Breton Minnehan, Andreas Savakis
Publikováno v:
CVPR
We propose a data-driven approach for deep convolutional neural network compression that achieves high accuracy with high throughput and low memory requirements. Current network compression methods either find a low-rank factorization of the features
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::6409cc4f5e957c20ec18e347b2fc5626
http://arxiv.org/abs/1903.04988
http://arxiv.org/abs/1903.04988
Publikováno v:
Sensors (Basel, Switzerland)
Sensors, Vol 20, Iss 2, p 547 (2020)
Sensors, Vol 20, Iss 2, p 547 (2020)
In recent years, deep learning-based visual object trackers have achieved state-of-the-art performance on several visual object tracking benchmarks. However, most tracking benchmarks are focused on ground level videos, whereas aerial tracking present
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::41f88cd1287be857aba9217908eef465
https://doi.org/10.3390/s20020547
https://doi.org/10.3390/s20020547