Michael Schaffner

Research

My research is centered around the design of hardware systems and VLSI architectures in the field of real-time signal and video processing.

LNS Arithmetic Units (LNUs)

Corresponding papers and theses:

A 65nm CMOS 6.4-to-29.2pJ/FLOP@0.8V Shared Logarithmic Floating Point Unit for Acceleration of Nonlinear Function Kernels in a Tightly Coupled Processor Cluster.
Michael Gautschi, Michael Schaffner, Frank K. Gürkaynak, Luca Benini. Proceedings of ISSCC , San Francisco, 2016. (publications)

High-Efficiency Logarithmic Number Unit Design based on an Improved Cotransformation Scheme.
Youri Popoff, Florian Scheidegger, Michael Schaffner, Michael Gautschi, Frank K. Gürkaynak, Luca Benini. Proceedings of DATE , Dresden, 2016. (publications)

Accuracy and Performance Trade-offs of Logarithmic Number Units in Multi-Core Clusters.
Michael Schaffner, Michael Gautschi, Frank K. Gürkaynak, Luca Benini. 23rd IEEE Symposium on Computer Arithmetic (ARITH 2016) , Santa Clara, USA, 2016. (publications)

An Extended Shared Logarithmic Unit for Nonlinear Function Kernel Acceleration in a 65-nm CMOS Multicore Cluster.
Michael Gautschi, Michael Schaffner, Frank K. Gürkaynak, Luca Benini. IEEE Journal of Solid-State Circuits, 52(1):98-112, 2017. (publications)

Energy-Efficient VLSI Architectures for Real-Time and 3D Video Processing.
Michael Stefano Fritz Schaffner Dissertation. ETH Zürich. 2017. No. 24261. (publications)

Real-Time Stereo to Multiview Conversion

Current three dimensional video systems have the drawback that the viewers must wear glasses in order to perceive depth effects. Furthermore, they cannot convey other depth cues such as motion parallax. A new generation of displays - so called multiview automultiscopic displays (MADs) - do not suffer from those problems. They project several images simultaneously into different spatial directions allowing the viewers to see two different images with both eyes. The viewers can also experience a limited motion parallax by moving around the different viewing zones. At present most three dimensional video footage is captured in stereoscopic 3D (S3D) format. So called multiview synthesis methods are employed to generate the large amount of views that are required to operate MADs. However, the number of views is highly display dependent, and it is infeasible to transmit several views to the display per frame. Thus it is desirable to have this functionality built into the displays themselves.

Links:
Our video on Youtube
IIS Projects Page
IBC 2015 Demo Page

Corresponding papers and theses:

A Multiview Synthesis Core in 65nm CMOS.
Michael Schaffner,
Master Thesis at the Integrated Systems Laboratory (IIS) at ETHZ, 2012.
(see also: Europractice Activity Report 2012)

Efficient Image Resampling for Multiview Displays.
Michael Schaffner, Pierre Greisen, Simon Heinzle, Aljoscha Smolic. Proceedings of the International Conference on Acoustics, Speech, and Signal Processing (ICASSP) , 2013. (publications)

A Real-Time 720p Feature Extraction Core Based on Semantic Kernels Binarized.
Michael Schaffner, Pascal Hager, Lukas Cavigelli, Pierre Greisen, Frank K. Gürkaynak, Hubert Kaeslin. Proceedings of the International Conference on Very Large Scale Integration (VLSI-SoC) , 2013. (publications)

MADmax: A 1080p Stereo-to-Multiview Rendering ASIC in 65 nm CMOS based on Image Domain Warping.
Michael Schaffner, Pierre Greisen, Simon Heinzle, Frank K. Gürkaynak, Hubert Kaeslin, Aljoscha Smolic. Proceedings of the European Solid-State Circuits Conference (ESSCIRC) , 2013. (publications)

An Approximate Computing Technique for Reducing the Complexity of a Direct-Solver for Sparse Linear Systems in Real-Time Video Processing.
Michael Schaffner, Frank K. Gürkaynak, Aljoscha Smolic, Hubert Kaeslin, Luca Benini. Proceedings of the Design Automation Conference (DAC) , San Francisco, USA, 2014. (publications)

A Complete Real-Time Feature Extraction and Matching System Based on Semantic Kernels Binarized.
Michael Schaffner, Pascal A. Hager, Lukas Cavigelli, Zhou Fang, Frank K. Gürkaynak, Pierre Greisen, Aljoscha Smolic, Hubert Kaeslin, Luca Benini. VLSI-SoC: At the Crossroads of Emerging Trends,, IFIP Advances in Information and Communication Technology. Springer Berlin Heidelberg, 2015. (publications)

DRAM or no-DRAM? Exploring Linear Solver Architectures for Image Domain Warping in 28 nm CMOS.
Michael Schaffner, Frank K. Gürkaynak, Aljoscha Smolic, Luca Benini. Proceedings of DATE 2015 , Grenoble, France, 2015. (publications)

Automatic Multiview Synthesis - Towards a Mobile System on a Chip.
Michael Schaffner, Frank K. Gürkaynak, Hubert Kaeslin, Luca Benini, Aljoscha Smolic. Proceedings of VCIP 2015 , Singapore, 2015. (publications)

Automatic Multiview Synthesis - Prototype Demo.
Michael Schaffner, Frank K. Gürkaynak, Hubert Kaeslin, Luca Benini, Aljoscha Smolic. Proceedings of VCIP 2015 , Singapore, 2015. (publications)

Hybrid ASIC/FPGA System for Fully Automatic Stereo-to-Multiview Conversion using IDW.
Michael Schaffner, Frank K. Gürkaynak, Pierre Greisen, Hubert Kaeslin, Luca Benini, Aljosa Smolic. IEEE Transactions on Circuits and Systems for Video Technology, 26(11):2093-2108, 2016. (publications)

Towards Edge-Aware Spatio-Temporal Filtering in Real-Time.
M. Schaffner, F. Scheidegger, L. Cavigelli, H. Kaeslin, L. Benini and A. Smolic. Accepted for publication in Trans. on Image Processing (TIP), 2017. (publications)

Energy-Efficient VLSI Architectures for Real-Time and 3D Video Processing.
Michael Stefano Fritz Schaffner Dissertation. ETH Zürich. 2017. No. 24261. (publications)

Video Retargeting, Depth Remapping

Corresponding papers and theses:

EWA Splatting Core for Video Retargeting and Depth Remapping.
Marian Runo, Michael Schaffner,
Semester Thesis at the Integrated Systems Laboratory (IIS) at ETHZ, 2010.

Analysis and VLSI Implementation of EWA Rendering for Real-Time HD Video Applications.
Pierre Greisen, Michael Schaffner, Simon Heinzle, Marian Runo, Aljosa Smolic, Andreas Burg, Hubert Kaeslin, Markus Gross. IEEE Transactions on Circuits and Systems for Video Technology, 22(11):1577-1589, 2012. (publications)

A General-Transformation EWA View Rendering Engine for 1080p Video in 130nm CMOS.
Pierre Greisen, Richard Emler, Michael Schaffner, Simon Heinzle, Frank Gürkaynak. IFIP/IEEE 20th International Conference on Very Large Scale Integration (VLSI-SoC 2012), Santa Cruz, USA, 2012.
(publications)

Other

Design and Evaluation of a Low-Power Sensor Device for Induced Rockfall Experiments.
A. Caviezel, M. Schaffner, L. Cavigelli, Y. Buehler, P. Bartelt, M. Magno, L. Benini. Accepted for publication in Transactions on Instrumentation and Measurement, 2017. (publications)

Impact of Temporal Subsampling on Accuracy and Performance in Practical Video Classification.
F. Scheidegger, L. Cavigelli, M. Schaffner, A. C. I. Malossi, C. Bekas, L. Benini. roc. of the European Signal Processing Conference (EUSIPCO) , Kos, Greece, 2017. (publications)

StoneNode: A Low-Power Sensor Device for Induced Rockfall Experiments.
P. Niklaus, T. Birchler, T. Aebi, M. Schaffner, L. Cavigelli, A. Caviezel, M. Magno, L. Benini. Proc. of the Sensors Applications Symposium (SAS) , Glassboro, USA, 2017. (publications)

Changing Camera Distance in a Single Image.
Michael Schaffner,
Semester Thesis at the Computer Vision Laboratory (BIWI) at ETHZ, 2011.

Realtime Vinyltracking on a DSP.
Samuel Gähwiler, Stefan Sommer, Michael Schaffner,
Bachelor Team Project at the Integrated Systems Laboratory (IIS) at ETHZ, 2009.

last update: July 11 2018 16:34:10.

Links:

ETH Zürich (ETHZ)
Disney Research Zurich (DRZ)
Integrated Systems Laboratory (IIS)
Microelectronics Design Center (DZ)
IIS Chipgallery