Search

Henrique S Malvar

from Kirkland, WA
Age ~66
Get Report
Also known as:
Henrique Sarmento Malvar
Phone and address:
2025 Rose Point Ln, Kirkland, WA 98033
(425) 558-9752
Related to:
Paul G Enriquez, 49Francisco Enriquez, 68
Connected to:
Grace L Malvar, 44Marvin Malvar, 47Sean Malvar, 35Victor Malvarez, 63Stanley B Malvarosa, 92Jose G Malvas, 57Analicia Z Malvasia, 76Steven R Malvasi, 31Christopher J Malvaso, 39Frank A Malvaso, 41

Henrique Malvar Phones & Addresses

  • 2025 Rose Point Ln, Kirkland, WA 98033 (425) 558-9752
  • 2302 233Rd Ave NE, Sammamish, WA 98074 (425) 836-5214
  • Redmond, WA
  • Bellevue, WA
  • Reno, NV
  • Kiona, WA
  • Cambridge, MA
  • Danvers, MA

Resumes

Resumes

Henrique Malvar Photo 1

Distinguished Engineer, Microsoft

View page
Location:
Seattle, WA
Industry:
Computer Software
Work:
Microsoft since Sep 2010
Chief Scientist
Microsoft since Mar 2006
Distinguished Engineer
University of Washington since 1999
Affiliate Professor
Microsoft Mar 2007 - Sep 2010
Managing Director, Microsoft Research Redmond
Microsoft Mar 2004 - Mar 2007
Director, Microsoft Research Redmond
Education:
Massachusetts Institute of Technology 1982 - 1986
Ph.D., Electrical Engineering and Computer Science Universidade Federal do Rio de Janeiro 1978 - 1979
M.Sc., Electrical Engineering Universidade de Brasília 1974 - 1977
B.S.E.E., Electrical Engineering
Skills:
Signal Processing
Algorithms
Image Processing
Digital Signal Processing
Digital Image Processing
Software Development
System Architecture
Strategy
Machine Learning
Audio Processing
Software Engineering
Big Data
Computer Science
Algorithm Design
Video Compression
Distributed Systems
Audio Codecs
Data Compression
Codecs
Information Retrieval
Computer Architecture
Artificial Intelligence
Neural Networks
C++
Cryptography
Software Design
C
Audio Signal Processing
Image Codecs
Cloud Computing
Awards:
Member of the Brazilian Academy of Sciences
Brazilian Academy of Sciences
"The Brazilian Academy of Sciences was founded in 1916 and brings together the most eminent scientists in the Mathematical, Physical, Chemical, Earth, Biological, Biomedical, Health, Agricultural, Engineering and Social Sciences."
Member of the U.S. National Academy of Engineering
U.S. National Academy of Engineering
For contributions to multiresolution signal processing and multimedia signal compression and standards. "Election to National Academy of Engineering membership is one of the highest professional honors accorded an engineer. Members have distinguished themselves in business and academic management, in technical positions, as university faculty, and as leaders in government and private engineering organizations."
Technical Achievement Award
IEEE Signal Processing Society
For development of lapped transforms and contributions to audio and video coding techniques. "The Technical Achievement Award honors a person who, over a period of years, has made outstanding technical contributions to theory and/or practice in technical areas within the scope of the Society, as demonstrated by publications, patents, or recognized impact on the field."
Fellow of the IEEE
Institute of Electrical and Electronics Engineers
For contributions to the theory and practice of lapped transforms, fast multirate filter banks, and signal coding. "IEEE Fellow is a distinction reserved for select IEEE members whose extraordinary accomplishments in any of the IEEE fields of interest are deemed fitting of this prestigious grade elevation."
Marconi Young Scientist Award
Marconi International Fellowship / Marconi Society
Awarded to a researcher / scientist in telecommunications, under the age of 27, in Brazil.
Henrique Malvar Photo 2

Henrique Malvar

View page
Skills:
Twilio
Software

Publications

Us Patents

Adaptive Filtering System And Method For Adaptively Canceling Echoes And Reducing Noise In Digital Signals

View page
US Patent:
6473409, Oct 29, 2002
Filed:
Feb 26, 1999
Appl. No.:
09/259390
Inventors:
Henrique S. Malvar - Redmond WA
Assignee:
Microsoft Corp. - Redmond WA
International Classification:
H04B 320
US Classification:
370286, 370203
Abstract:
The present invention is embodied in a system and method for performing spectral analysis of a digital signal having a discrete duration by spectrally decomposing the digital signal at predefined frequencies uniformly distributed over a sampling frequency interval into complex frequency coefficients so that magnitude and phase information at each frequency is immediately available to produce a modulated complex lapped transform (MCLT). The system includes real and imaginary window processors and real and imaginary transform processors. The real and imaginary window processors receive the input signal and apply and compute butterfly coefficients for the real and imaginary parts of the signal to produce resulting real and imaginary vectors, respectively. The real and imaginary transform processors compute spatial transforms on the real and imaginary vectors to produce real and imaginary transform coefficient of the MCLT, respectively. The MCLT is a biorthogonal spectral transformation system, in the sense that the original time domain signal can be reconstructed exactly by an inverse MCLT operator.

Lossless Adaptive Encoding Of Finite Alphabet Data

View page
US Patent:
6477280, Nov 5, 2002
Filed:
Mar 26, 1999
Appl. No.:
09/277255
Inventors:
Henrique S. Malvar - Redmond WA
Assignee:
Microsoft Corporation - Redmond WA
International Classification:
G06K 936
US Classification:
382245, 382232
Abstract:
An encoder reorders quantized wavelet coefficients to cluster large and small wavelet coefficients into separate groups without requiring the use of data-dependent data structures. The coefficients are then adaptively encoded based on a run-length code which continuously modifies a parameter that controls the codewords uses to represent strings of quantized coefficients, seeking to minimize the number of bits spent in the codewords. A matrix of indices contains the coarsest coefficients in the upper left corner, and filling in low high and high low sub bands in larger and larger blocks in an alternating manner, such that low high sub bands comprise the top of the matrix and the high low sub bands comprise the left side of the matrix. The shortest codewords are assigned to represent a run of the most likely character having length of 2 , where k is a parameter. k is adjusted based on successive characters being encountered.

System And Method For Producing Modulated Complex Lapped Transforms

View page
US Patent:
6487574, Nov 26, 2002
Filed:
Feb 26, 1999
Appl. No.:
09/259669
Inventors:
Henrique S. Malvar - Redmond WA
Assignee:
Microsoft Corp. - Redmond WA
International Classification:
G06F 1714
US Classification:
708400, 708402
Abstract:
The present invention is embodied in a system and method for performing spectral analysis of a digital signal having a discrete duration by spectrally decomposing the digital signal at predefined frequencies uniformly distributed over a sampling frequency interval into complex frequency coefficients so that magnitude and phase information at each frequency is immediately available to produce a modulated complex lapped transform (MCLT). The system includes real and imaginary window processors and real and imaginary transform processors. The real and imaginary window processors receive the input signal and apply and compute butterfly coefficients for the real and imaginary parts of the signal to produce resulting real and imaginary vectors, respectively. The real and imaginary transform processors compute spatial transforms on the real and imaginary vectors to produce real and imaginary transform coefficient of the MCLT, respectively. The MCLT is a biorthogonal spectral transformation system, in the sense that the original time domain signal can be reconstructed exactly by an inverse MCLT operator.

Modulated Complex Lapped Transform For Integrated Signal Enhancement And Coding

View page
US Patent:
6496795, Dec 17, 2002
Filed:
May 5, 1999
Appl. No.:
09/305690
Inventors:
Henrique S. Malvar - Redmond WA
Assignee:
Microsoft Corporation - Redmond WA
International Classification:
G10L 1502
US Classification:
704203, 708400
Abstract:
The present invention is embodied in a system and method for performing spectral analysis of a digital signal having a discrete duration by spectrally decomposing the digital signal at predefined frequencies uniformly distributed over a sampling frequency interval into complex frequency coefficients so that magnitude and phase information at each frequency is immediately available to produce a modulated complex lapped transform (MCLT). The present invention includes a MCLT processor, an acoustic echo cancellation device and a noise reducer integrated with an encoder/decoder device.

Compression Of Animated Geometry Using Basis Decomposition

View page
US Patent:
6606095, Aug 12, 2003
Filed:
Jun 7, 1999
Appl. No.:
09/326969
Inventors:
Jerome E. Lengyel - Seattle WA
Brian K. Guenter - Redmond WA
Henrique Sarmento Malvar - Redmond WA
Assignee:
Microsoft Corporation - Redmond WA
International Classification:
G06T 1700
US Classification:
345473
Abstract:
Methods for coding a time-dependent geometry stream include a basis decomposition coder and a column/row prediction coder. The basis decomposition coder uses principal component analysis to decompose a time dependent geometry matrix into basis vectors and weights. The weights and basis vectors are coded separately. Optionally, the residual between a mesh constructed from the weights and basis vectors and the original mesh can be encoded as well. The column/row predictor exploits coherence in a matrix of time dependent geometry by encoding differences among neighboring rows and columns. Row and column sorting optimizes this form of coding by re-arranging rows and columns to improve similarity among neighboring rows/columns.

Reordering Wavelet Coefficients For Improved Encoding

View page
US Patent:
6678419, Jan 13, 2004
Filed:
Mar 26, 1999
Appl. No.:
09/280135
Inventors:
Henrique S. Malvar - Redmond WA
Assignee:
Microsoft Corporation - Redmond WA
International Classification:
G06K 936
US Classification:
382240, 37524008, 37524026
Abstract:
An encoder reorders quantized wavelet coefficients to cluster large and small wavelet coefficients into separate groups without requiring the use of data-dependent data structures. The coefficients are then adaptively encoded based on a run-length code which continuously modifies a parameter that controls the codewords uses to represent strings of quantized coefficients, seeking to minimize the number of bits spent in the codewords. A matrix of indices contains the coarsest coefficients in the upper left corner, and filling in low high and high low sub bands in larger and larger blocks in an alternating manner, such that low high sub bands comprise the top of the matrix and the high low sub bands comprise the left side of the matrix. The shortest codewords are assigned to represent a run of the most likely character having length of 2 , where k is a parameter. k is adjusted based on successive characters being encountered.

System And Method For Trainable Nonlinear Prediction Of Transform Coefficients In Data Compression

View page
US Patent:
6704718, Mar 9, 2004
Filed:
Jun 5, 2001
Appl. No.:
09/681789
Inventors:
Chris J. C. Burges - Bellevue WA
Patrice Y. Simard - Bellevue WA
Henrique S. Malvar - Bellevue WA
Assignee:
Microsoft Corporation - Redmond WA
International Classification:
G06E 100
US Classification:
706 21, 706 15, 706 16
Abstract:
A system and method for performing trainable nonlinear prediction of transform coefficients in data compression such that the number of bits required to represent the data is reduced. The nonlinear prediction data compression system includes a nonlinear predictor for generating predicted transform coefficients, a nonlinear prediction encoder that uses the predicted transform coefficients to encode original data, and a nonlinear prediction decoder that uses the predicted transform coefficients to decode the encoded bitstream and reconstruct the original data. The nonlinear predictor may be trained using training techniques, including a novel in-loop training technique of the present invention. The present invention also includes a method for using a nonlinear predictor to encode and decode data. The method also includes improving the performance of the nonlinear prediction data compression and decompression using several novel speedup techniques.

Watermark Detection Via Cardinality-Scaled Correlation

View page
US Patent:
6738744, May 18, 2004
Filed:
Dec 8, 2000
Appl. No.:
09/733576
Inventors:
Darko Kirovski - Redmond WA
Henrique Malvar - Redmond WA
Assignee:
Microsoft Corporation - Redmond WA
International Classification:
G10L 1522
US Classification:
704273, 713193
Abstract:
Described herein is an audio watermarking technology for detecting watermarks in audio signals, such as a music clip. The watermark identifies the content producer, providing a signature that is embedded in the audio signal and cannot be removed. The watermark is designed to survive all typical kinds of processing and all types of malicious attacks that attempt to remove or modify the watermark from the signal. The implementations of the watermark detecting system, described herein, support quick, efficient, and accurate detection of watermarks by the specifically designed watermark detecting system. In one described implementation, a watermark detecting system employs an improved normalized covariance test to determine the presence of a watermark using less expensive materials (hardware), quicker calculations, and a more accurate test (than the original correlation test). In other described implementations, a watermark detecting system employs a cepstrum filter and dynamic processing to minimize the affect of the ânoiseâ in the watermarked signal. The ânoiseâ is the original content of the signal before such signal was watermarked.

Isbn (Books And Publications)

Signal Processing With Lapped Transforms

View page
Author

Henrique S. Malvar

ISBN #

0890064679

Control profile
Henrique S Malvar from Kirkland, WA, age ~66 Get Report