منابع مشابه
Super Efficiency in Vector Optimization
We introduce a new concept of efficiency in vector optimization. This concept, super efficiency, is shown to have many desirable properties. In particular, we show that in reasonable settings the super efficient points of a set are norm-dense in the efficient frontier. We also provide a Chebyshev characterization of super efficient points for nonconvex sets and a scalarization theory when the u...
متن کاملImage Classification Using Super-Vector Coding of Local Image Descriptors
This paper introduces a new framework for image classification using local visual descriptors. The pipeline first performs a nonlinear feature transformation on descriptors, then aggregates the results together to form image-level representations, and finally applies a classification model. For all the three steps we suggest novel solutions which make our approach appealing in theory, more scal...
متن کاملA Hybrid Vector Wiener Filter Approach to Translational Super-Resolution
We address the problem of purely-translational super-resolution (SR) for signals in arbitrary dimensions. We show that discretization, a key step in many SR algorithms, inevitably leads to inaccurate modeling. Instead, we treat the problem entirely in the continuous domain by modeling the signal as a continuous-space random process and deriving its linear minimum mean-squared error (LMMSE) esti...
متن کاملInterpolation Based Image Super Resolution by Support-Vector-Regression
The higher resolution image can be reconstructed from lower resolution images using SuperResolution (SR) algorithm based on Support Vector Regression (SVR) by combining the pixel intensity values with local gradient information. Support Vector Machine (SVM) can construct a hyperplane in a high or infinite dimensional space which can be used for classification. Its regression version, Support Ve...
متن کاملHybrid Super Vector with Improved Dense Trajectories for Action Recognition
With recent improved dense trajectory features (HOG, warped HOF, and warped MBH), we employ two advanced super vector methods, namely Fisher Vector (FV) and soft Vector of Locally Aggregated Descriptors (VLAD-K) to encode them separately. The two individual super vectors are concatenated into a Hybrid Super Vector, and a linear SVM classifier is used to predict labels. We achieve 87.46%1 in ave...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: La Revue pour l’histoire du CNRS
سال: 2009
ISSN: 1298-9800,1955-2408
DOI: 10.4000/histoire-cnrs.9061