Montag, 29. Mai 2017

[HIForum] [Kolloquium] REMINDER: Informatics Colloquium Mo 29.05.17 at 17:15 in B-201 with Prof. Dr.-Ing. Kai Uwe Barthel/ HTW Berlin

Dear all,


May we kindly remind you of the today's talk at 17:15 in room B-201. Please find the details below.

 

Best regards


On behalf of the colloquium committee

Stephanie Schulte Hemming

Universität Hamburg

 

 

Von: Fbi-alle <fbi-alle-bounces@mailhost.informatik.uni-hamburg.de> im Auftrag von "Schulte Hemming, Stephanie" <schulte@informatik.uni-hamburg.de>
Datum: Dienstag, 16. Mai 2017 um 09:47
An: "kolloquium@informatik.uni-hamburg.de" <kolloquium@informatik.uni-hamburg.de>
Betreff: [Fbi-alle] [Kolloquium] INVITATION: Informatics Colloquium Mo 29.05.17 at 17:15 in B-201 with Prof. Dr.-Ing. Kai Uwe Barthel/ HTW Berlin

 

This is an invitation to the next Informatics Colloquium on Monday, 29 May 2017, 17:15, Campus "Informatikum/Stellingen", Room B-201 (Konrad-Zuse lecture hall). The talk entitled "Visually Browsing Millions of Images using Image Graphs" will be held by Prof. Dr.-Ing. Kai Uwe Barthel, Hochschule für Technik und Wirtschaft, HTW Berlin.

                

This talk will be held in English. The colloquium committee is looking forward to seeing you all there and to sharing this talk with you. For details on the series of colloquiums planned, please visit  https://www.inf.uni-hamburg.de/home/kolloquium/sose17.html

 

On behalf of the colloquium committee

Stephanie Schulte Hemming

Universität Hamburg

 

 

ABSTRACT:

In the past an efficient and satisfactory image search was only possible by using a combation of keywords and low-level visual image features. Recently Convolutional Neural Networks (CNNs) have enabled automatic understanding of images. This results in a multitude of new applications and improved visual image search systems. This talk provides an overview of the different methods for image search, gives an overview of the principle of CNNs and shows how future image search systems could look like. We present a new approach to visually explore very large sets of untagged images. High quality image descriptors are generated using transformed activations of a convolutional neural network. These features are used to model image similarities, from which a hierarchical image graph is build. We show how such a graph can be constructed efficiently. Best user experience for navigating this graph is achieved by projecting sub-graphs onto a regular 2D-image map. This allows users to explore the image graph similar to navigation services.

 

BIO:

Prof. Dr. Kai Uwe Barthel studied Electrical Engineering at the Technical University of Berlin. At the university's department for image and signal processing he finished in 1996 his PhD thesis about fractal image compression. In 1998 he became head of R&D with LuraTech Inc. where hard- and software for image compression were developed. He also participated in the JPEG2000 standardization committee. His research topic of that time was the segmentation of documents for Mixed-Raster-Content compression. In 2001 he became a professor for visual computing at HTW Berlin, University of Applied Sciences. At HTW Berlin he is teaching courses such as image processing, computer vision, visual information retrieval, and machine learning. Main research topics are visual image search and automatic image understanding. In 2009 he founded pixolution, a company for visual image search. His latest research includes automatic keywording of images, image clustering and the development of visual image navigation systems such as www.picsbuffet.com

  

CONTACT:

Prof. Dr.-Ing. H.-Siegfried Stiehl, BV/HCC

 

 

 

 

 

Montag, 22. Mai 2017

[HIForum] [Kolloquium] REMINDER: Informatics Colloquium Mo 22 May 2017, 17:15, B-201 "Signal Processing for Speech Enhancement" by Prof. Dr. Timo Gerkmann

_______________________________________________
Kolloquium mailing list
Kolloquium@mailhost.informatik.uni-hamburg.de
https://mailhost.informatik.uni-hamburg.de/mailman/listinfo/kolloquium_______________________________________________
Dear all,

May we kindly remind you of the 
today's inaugural lecture with Prof. Timo Gerkmann at 17:15 in room B-201. Please find the details below.

Best regards

On behalf of the colloquium committee
Stephanie Schulte Hemming
Universität Hamburg


Von: Fbi-alle [fbi-alle-bounces@mailhost.informatik.uni-hamburg.de]" im Auftrag von "Schulte Hemming, Stephanie [schulte@informatik.uni-hamburg.de]
Gesendet: Dienstag, 9. Mai 2017 09:10
An: kolloquium@informatik.uni-hamburg.de
Betreff: [Fbi-alle] [Kolloquium] INVITATION: Informatics Colloquium Mo 22 May 2017, 17:15, B-201 "Signal Processing for Speech Enhancement" by Prof. Dr. Timo Gerkmann

This is an invitation to the next Informatics Colloquium on Monday, 22 May 2017, 17:15 s.t., Campus "Informatikum/Stellingen", Room B-201 (Konrad-Zuse lecture hall). The talk entitled "Signal Processing for Speech Enhancement" will be held by Prof. Dr. Timo Gerkmann, since August 2016 Professor of Signal Processing at Universität Hamburg, department of informatics.
               
This talk will be held in English. The colloquium committee is looking forward to seeing you all there and to sharing this talk with you. For details on the series of colloquiums planned, please visit  https://www.inf.uni-hamburg.de/home/kolloquium/sose17.html
 
On behalf of the colloquium committee
Stephanie Schulte Hemming
Universität Hamburg
 
 
ABSTRACT:
Speech is arguably the most natural and important means for human communication. In recent years, speech has become more and more feasible and important also for human-computer interaction. As many speech communications devices like smartphones and hearing aids are portable or even wearable, they are frequently used in very noisy environments such as crowded restaurants, busy streets, or in a cafeteria. However, the noise and reverberation in such environments may make speech communication difficult or even impossible.

In the Signal Processing (SP) group, we focus on audio signal processing and in particular aim at making speech communication work more robustly in noisy and reverberant environments. To achieve this, we combine prior knowledge about the signal (e.g. speech), the environment (e.g. the room) and the sink (e.g. the human ear) with rigorous mathematical optimization procedures. In this talk, we will introduce the general concepts used for signal enhancement and highlight our recent contributions.

 
BIO:
* since September 2016: Professor of Signal Processing, Universität Hamburg
* 2015-2016: Principal Scientist Audio & Acoustics, Technicolor Research & Innovation
* 2011-2015: Juniorprofessor of Speech Signal Processing, Cluster of Excellence Hearing4all, Universität Oldenburg
* 2010-2011: Postdoctoral Researcher, Sound and Image Proc. Lab, KTH, Royal Institute of Technology, Stockholm
* 2005-2010: PhD Student, Institute of Communication Acoustics, Ruhr-Universität Bochum
* 2005: Siemens Corporate Research, Princeton, NJ, USA
 
CONTACT:
Prof. Dr. Tilo Böhmann, ITMC


Dienstag, 16. Mai 2017

[HIForum] [Kolloquium] INVITATION: Informatics Colloquium Mo 29.05.17 at 17:15 in B-201 with Prof. Dr.-Ing. Kai Uwe Barthel/ HTW Berlin

This is an invitation to the next Informatics Colloquium on Monday, 29 May 2017, 17:15, Campus "Informatikum/Stellingen", Room B-201 (Konrad-Zuse lecture hall). The talk entitled "Visually Browsing Millions of Images using Image Graphs" will be held by Prof. Dr.-Ing. Kai Uwe Barthel, Hochschule für Technik und Wirtschaft, HTW Berlin.

                

This talk will be held in English. The colloquium committee is looking forward to seeing you all there and to sharing this talk with you. For details on the series of colloquiums planned, please visit  https://www.inf.uni-hamburg.de/home/kolloquium/sose17.html

 

On behalf of the colloquium committee

Stephanie Schulte Hemming

Universität Hamburg

 

 

ABSTRACT:

In the past an efficient and satisfactory image search was only possible by using a combation of keywords and low-level visual image features. Recently Convolutional Neural Networks (CNNs) have enabled automatic understanding of images. This results in a multitude of new applications and improved visual image search systems. This talk provides an overview of the different methods for image search, gives an overview of the principle of CNNs and shows how future image search systems could look like. We present a new approach to visually explore very large sets of untagged images. High quality image descriptors are generated using transformed activations of a convolutional neural network. These features are used to model image similarities, from which a hierarchical image graph is build. We show how such a graph can be constructed efficiently. Best user experience for navigating this graph is achieved by projecting sub-graphs onto a regular 2D-image map. This allows users to explore the image graph similar to navigation services.

 

BIO:

Prof. Dr. Kai Uwe Barthel studied Electrical Engineering at the Technical University of Berlin. At the university's department for image and signal processing he finished in 1996 his PhD thesis about fractal image compression. In 1998 he became head of R&D with LuraTech Inc. where hard- and software for image compression were developed. He also participated in the JPEG2000 standardization committee. His research topic of that time was the segmentation of documents for Mixed-Raster-Content compression. In 2001 he became a professor for visual computing at HTW Berlin, University of Applied Sciences. At HTW Berlin he is teaching courses such as image processing, computer vision, visual information retrieval, and machine learning. Main research topics are visual image search and automatic image understanding. In 2009 he founded pixolution, a company for visual image search. His latest research includes automatic keywording of images, image clustering and the development of visual image navigation systems such as www.picsbuffet.com

  

CONTACT:

Prof. Dr.-Ing. H.-Siegfried Stiehl, BV/HCC

 

 

 

 

 

Dienstag, 9. Mai 2017

[HIForum] [Kolloquium] INVITATION: Informatics Colloquium Mo 22 May 2017, 17:15, B-201 "Signal Processing for Speech Enhancement" by Prof. Dr. Timo Gerkmann

This is an invitation to the next Informatics Colloquium on Monday, 22 May 2017, 17:15 s.t., Campus "Informatikum/Stellingen", Room B-201 (Konrad-Zuse lecture hall). The talk entitled "Signal Processing for Speech Enhancement" will be held by Prof. Dr. Timo Gerkmann, since August 2016 Professor of Signal Processing at Universität Hamburg, department of informatics.
               
This talk will be held in English. The colloquium committee is looking forward to seeing you all there and to sharing this talk with you. For details on the series of colloquiums planned, please visit  https://www.inf.uni-hamburg.de/home/kolloquium/sose17.html
 
On behalf of the colloquium committee
Stephanie Schulte Hemming
Universität Hamburg
 
 
ABSTRACT:
Speech is arguably the most natural and important means for human communication. In recent years, speech has become more and more feasible and important also for human-computer interaction. As many speech communications devices like smartphones and hearing aids are portable or even wearable, they are frequently used in very noisy environments such as crowded restaurants, busy streets, or in a cafeteria. However, the noise and reverberation in such environments may make speech communication difficult or even impossible.

In the Signal Processing (SP) group, we focus on audio signal processing and in particular aim at making speech communication work more robustly in noisy and reverberant environments. To achieve this, we combine prior knowledge about the signal (e.g. speech), the environment (e.g. the room) and the sink (e.g. the human ear) with rigorous mathematical optimization procedures. In this talk, we will introduce the general concepts used for signal enhancement and highlight our recent contributions.
 
BIO:
* since September 2016: Professor of Signal Processing, Universität Hamburg
* 2015-2016: Principal Scientist Audio & Acoustics, Technicolor Research & Innovation
* 2011-2015: Juniorprofessor of Speech Signal Processing, Cluster of Excellence Hearing4all, Universität Oldenburg
* 2010-2011: Postdoctoral Researcher, Sound and Image Proc. Lab, KTH, Royal Institute of Technology, Stockholm
* 2005-2010: PhD Student, Institute of Communication Acoustics, Ruhr-Universität Bochum
* 2005: Siemens Corporate Research, Princeton, NJ, USA
 
CONTACT:
Prof. Dr. Tilo Böhmann, ITMC

---------------------------------------

Please note:   
Due to a current design update of our website in Fiona, which is about to be implemented, this talk will not appear on the colloquium website until tomorrow.