The Experts below are selected from a list of 282 Experts worldwide ranked by ideXlab platform
Francois Bremond - One of the best experts on this subject based on the ideXlab platform.
-
ECCV Workshops (1) - Mitigating Bias in Gender, Age and Ethnicity Classification: a Multi-Task Convolution Neural Network Approach
Lecture Notes in Computer Science, 2019Co-Authors: Abhijit Das, Antitza Dantcheva, Francois BremondAbstract:This work explores joint classification of gender, age and race. Specifically, we here propose a Multi-Task Convolution Neural Network (MTCNN) employing joint Dynamic Loss weight adjustment towards classification of named soft biometrics, as well as towards mitigation of soft biometrics related bias. The proposed algorithm achieves promising results on the UTKFace and the Bias Estimation in Face Analytics (BEFA) datasets and was ranked first in the BEFA Challenge of the European Conference of Computer Vision (ECCV) 2018.
-
Mitigating Bias in Gender, Age and Ethnicity Classification: a Multi-Task Convolution Neural Network Approach
2018Co-Authors: Abhijit Das, Antitza Dantcheva, Francois BremondAbstract:This work explores joint classification of gender, age and race. Specifically, we here propose a Multi-Task Convolution Neural Network (MTCNN) employing joint Dynamic Loss weight adjustment towards classification of named soft biometrics, as well as towards mitigation of soft biometrics related bias. The proposed algorithm achieves promising results on the UTKFace and the Bias Estimation in Face Analytics (BEFA) datasets and was ranked first in the the BEFA Challenge of the European Conference of Computer Vision (ECCV) 2018.
Sergei Gudojnikov - One of the best experts on this subject based on the ideXlab platform.
-
impact of moisture content on Dynamic mechanical properties and transition temperatures of wood
Wood Material Science and Engineering, 2017Co-Authors: Oleg V Startsev, Alexey Makhonkov, Vladimir Erofeev, Sergei GudojnikovAbstract:AbstractThe Dynamic shear modulus and the Loss modulus of Betula alba, Ulmus parvifolia, Quercus robur, Acer platanoides, Tilia cordata, Fraxinus excelsior and Pinus sylvestris wood were measured using an inverted torsion pendulum within a wide temperature range. The glass transition temperature of the lignin–carbohydrate complex and the decomposition temperature of the wood cellulose were estimated. The temperature band from 170°C to 240°С shows the transition of the lignin–cellulose complex from the glassy to the rubbery state. Mechanical properties of different types of wood are affected by moisture and anatomical differences, but glass transition and decomposition temperatures are the same. More than 5% of moisture in the wood stored at normal conditions were found. After drying, the increase of Dynamic shear modulus of wood over the entire region of the glassy state was observed. The intensity of maximum peak of Dynamic Loss modulus is also increased due to activation of the segmental motion of macro...
Abhijit Das - One of the best experts on this subject based on the ideXlab platform.
-
ECCV Workshops (1) - Mitigating Bias in Gender, Age and Ethnicity Classification: a Multi-Task Convolution Neural Network Approach
Lecture Notes in Computer Science, 2019Co-Authors: Abhijit Das, Antitza Dantcheva, Francois BremondAbstract:This work explores joint classification of gender, age and race. Specifically, we here propose a Multi-Task Convolution Neural Network (MTCNN) employing joint Dynamic Loss weight adjustment towards classification of named soft biometrics, as well as towards mitigation of soft biometrics related bias. The proposed algorithm achieves promising results on the UTKFace and the Bias Estimation in Face Analytics (BEFA) datasets and was ranked first in the BEFA Challenge of the European Conference of Computer Vision (ECCV) 2018.
-
Mitigating Bias in Gender, Age and Ethnicity Classification: a Multi-Task Convolution Neural Network Approach
2018Co-Authors: Abhijit Das, Antitza Dantcheva, Francois BremondAbstract:This work explores joint classification of gender, age and race. Specifically, we here propose a Multi-Task Convolution Neural Network (MTCNN) employing joint Dynamic Loss weight adjustment towards classification of named soft biometrics, as well as towards mitigation of soft biometrics related bias. The proposed algorithm achieves promising results on the UTKFace and the Bias Estimation in Face Analytics (BEFA) datasets and was ranked first in the the BEFA Challenge of the European Conference of Computer Vision (ECCV) 2018.
Oleg V Startsev - One of the best experts on this subject based on the ideXlab platform.
-
impact of moisture content on Dynamic mechanical properties and transition temperatures of wood
Wood Material Science and Engineering, 2017Co-Authors: Oleg V Startsev, Alexey Makhonkov, Vladimir Erofeev, Sergei GudojnikovAbstract:AbstractThe Dynamic shear modulus and the Loss modulus of Betula alba, Ulmus parvifolia, Quercus robur, Acer platanoides, Tilia cordata, Fraxinus excelsior and Pinus sylvestris wood were measured using an inverted torsion pendulum within a wide temperature range. The glass transition temperature of the lignin–carbohydrate complex and the decomposition temperature of the wood cellulose were estimated. The temperature band from 170°C to 240°С shows the transition of the lignin–cellulose complex from the glassy to the rubbery state. Mechanical properties of different types of wood are affected by moisture and anatomical differences, but glass transition and decomposition temperatures are the same. More than 5% of moisture in the wood stored at normal conditions were found. After drying, the increase of Dynamic shear modulus of wood over the entire region of the glassy state was observed. The intensity of maximum peak of Dynamic Loss modulus is also increased due to activation of the segmental motion of macro...
Shanhui Fan - One of the best experts on this subject based on the ideXlab platform.
-
stopping and time reversing a light pulse using Dynamic Loss tuning of coupled resonator delay lines
Optics Letters, 2007Co-Authors: Sunil Sandhu, Michelle L Povinelli, Shanhui FanAbstract:We introduce a light-stopping process that uses Dynamic Loss tuning in coupled-resonator delay lines. We demonstrate via numerical simulations that increasing the Loss of selected resonators traps light in a zero group velocity mode concentrated in the low-Loss portions of the delay line. The large Dynamic range achievable for Loss modulation should increase the light-stopping bandwidth relative to previous approaches based on refractive index tuning.
-
Loss-Tuning of Coupled-Resonator Delay Lines Allows Light-Stopping of Large Bandwidth Signal
Integrated Photonics and Nanophotonics Research and Applications Slow and Fast Light, 2007Co-Authors: Sunil Sandhu, Michelle L Povinelli, Shanhui FanAbstract:We introduce a novel light-stopping process using Dynamic Loss tuning. The system allows a ~1THz bandwidth signal to be delayed for ~12ps. We present an analysis of the system during the Dynamic Loss tuning.
