Publications associées à Méthode de Bartlett

Robustness via Cross-Domain Ensembles

Amir Roshan Zamir, Oguzhan Fatih Kar, Shuqing Teresa Yeo

We present a method for making neural network predictions robust to shifts from the training data distribution. The proposed method is based on making predictions via a diverse set of cues (called 'middle domains') and ensembling them into one strong predi ...

IEEE2021

Visual Focus of Attention Estimation in 3D Scene with an Arbitrary Number of Targets

Jean-Marc Odobez, Rémy Alain Siegfried

Visual Focus of Attention (VFOA) estimation in conversation is challenging as it relies on difficult to estimate information (gaze) combined with scene features like target positions and other contextual information (speaking status) allowing to disambigua ...

IEEE COMPUTER SOC2021

Unpaired Cross-Spectral Pedestrian Detection Via Adversarial Feature Learning

Seungryong Kim

Even though there exist significant advances in recent studies, existing methods for pedestrian detection still have shown limited performances under challenging illumination conditions especially at nighttime. To address this, cross-spectral pedestrian de ...

IEEE2019

Two-step phase shifting interferometry based on orientation selective monogenic filtering

Pramod Rastogi, Rishikesh Dilip Kulkarni

We propose a phase step estimation algorithm in a two-step phase shifting interferometry setup based on orientation selective adaptive monogenic filtering (AMF). Initially, the AMF algorithm is used for the fringe pattern normalization. A minor modificatio ...

2019

Robust heart rate estimation using wrist-type photoplethysmographic signals during physical exercise: an approach based on adaptive filtering

Jean-Marc Vesin, Sibylle Fallet

Photoplethysmographic (PPG) signals are easily corrupted by motion artifacts when the subjects perform physical exercise. This paper introduces a two-step processing scheme to estimate heart rate (HR) from wrist-type PPG signals strongly corrupted by motio ...

Institute of Physics2017

Real-time multi-signal frequency tracking with a bank of notch filters to estimate the respiratory rate from the ECG

Jean-Marc Vesin, Leila Mirmohamadsadeghi

Measuring the instantaneous frequency of a signal rapidly and accurately is essential in many applications. However, the instantaneous frequency by definition is a parameter difficult to determine. Fourier-based methods introduce estimation delays as compu ...

Institute of Physics2016

Estimation of Phase and Its Higher Order Derivatives from a Single Complex Interferogram

Pramod Rastogi, Rishikesh Dilip Kulkarni

This paper reports a method for the simultaneous estimation of unwrapped phase and higher order phase derivatives from a single phase fringe pattern recorded in an optical interferometric setup, thereby overcoming substantial barriers to achieving such mea ...

2016

A Spectral Method for Generating Surrogate Graph Signals

Dimitri Nestor Alice Van De Ville, Elvira Pirondini, Martina Coscia, Anna Vybornova

The increasing availability of network data is leading to a growing interest in processing of signals on graphs. One notable tool for extending conventional signal-processing operations to networks is the graph Fourier transform that can be obtained as the ...

IEEE Institute of Electrical and Electronics Engineers2016

Non-Parametric Estimation of Surface Temperature of Li-ion Cells using Thermal Impulse Response

Mario Paolone, Dimitri Torregrossa, Ying Xiao

This paper focuses on the prediction of temperature profiles on the surface of lithium-ion cells using a non-parametric method. In particular, this paper proposes the adoption of the impulse response technique to compute and interpret cell surface temperat ...

2016

Three-dimensional displacement measurement from phase signals embedded in a frame in digital holographic interferometry

Pramod Rastogi, Rishikesh Dilip Kulkarni

A novel technique is proposed for the simultaneous measurement of all three components of displacement from a single recording of the interference field in digital holographic interferometry. The interference field is divided into a number of rectangular s ...

Optical Soc Amer2015

Matrix pencil based phase derivative estimation in digital holographic interferometry

Pramod Rastogi, Rishikesh Dilip Kulkarni

A novel technique is proposed for the direct two-dimensional phase derivatives estimation based on the matrix pencil approach in digital holographic interferometry. The interference field is represented as a two-dimensional complex sinusoidal signal in an ...

Iop Publishing Ltd2015

Iterative signal separation based multiple phase estimation in digital holographic interferometry

Pramod Rastogi, Rishikesh Dilip Kulkarni

We propose a new method for signal separation from a multicomponent interference field recorded in a digital holographic interferometry setup. The setup consisting of multiple object illuminating beams results in an interference field containing multiple s ...

Optical Society of America2015

Phase derivative estimation from a single interferogram using a Kalman smoothing algorithm

Pramod Rastogi, Rishikesh Dilip Kulkarni

We report a technique for direct phase derivative estimation from a single recording of a complex interferogram. In this technique, the interference field is represented as an autoregressive model with spatially varying coefficients. Estimates of these coe ...

Optical Society of America2015

Digital holographic moiré for the direct and simultaneous estimation of strain and slope fields

Pramod Rastogi, Rishikesh Dilip Kulkarni

A novel method is proposed for the direct and simultaneous estimation of multiple phase derivatives corresponding to strain and slope fields from a single moiré fringe pattern in digital holographic moiré. The interference field in a given row/column is a ...

Optical Society of America2014

Estimation of phase derivatives using discrete energy separation algorithm in digital holographic interferometry

Pramod Rastogi, Rishikesh Dilip Kulkarni

This Letter proposes a new method for the estimation of the first-and second-order phase derivatives corresponding to strain and curvature from a single fringe pattern in digital holographic interferometry. The method is based on a discrete energy separati ...

Optical Soc Amer2014

Direct phase derivative estimation using difference equation modeling in holographic interferometry

Pramod Rastogi, Rishikesh Dilip Kulkarni

A new method is proposed for the direct phase derivative estimation from a single spatial frequency modulated carrier fringe pattern in holographic interferometry. The fringe intensity in a given row/column is modeled as a difference equation of intensity ...

Iop Publishing Ltd2014

Simultaneous estimation of phase and phase derivative using a difference equation representation of the interference field

Pramod Rastogi, Rishikesh Dilip Kulkarni

A computationally efficient technique for fringe analysis in digital holographic interferometry using a difference equation representation of the interference field is presented. The spatially varying coefficient of the difference equation is estimated acc ...

Optical Soc Amer2014

Phase estimation using a state-space approach based method

Pramod Rastogi, Rajshekhar Gannavarpu

The paper demonstrates a phase estimation method in fringe analysis. The proposed method relies on local polynomial phase approximation and subsequent state-space formulation. The polynomial approximation of phase transforms phase extraction into a paramet ...

Elsevier2013

Multiple signal classification technique for phase estimation from a fringe pattern

Pramod Rastogi, Rajshekhar Gannavarpu

The paper introduces a multiple signal classification technique based method for fringe analysis. In the proposed method, the phase of a fringe pattern is locally approximated as a polynomial. The polynomial phase signal is then transformed to obtain signa ...

Optical Society of America2012

Application of the Matrix Pencil Method to Rational Fitting of Frequency-Domain Responses

Mario Paolone, Farhad Rachidi-Haeri, Hamidreza Karami, Keyhan Sheshyekani

This paper presents a general methodology based on the matrix pencil method (MPM) for the fitting of frequency-domain responses in order to be properly represented in time-domain analysis. By virtue of the proposed method, a rational fitting of the frequen ...

Ieee-Inst Electrical Electronics Engineers Inc2012