March 8th 2016

Keynote : Calin Cascaval, Qualcomm Symphony: Orchestrating Heterogeneity for Power Aware Computing.

February 16th 2016

The program is online.

January 27th 2016

The submission deadline is extended to Feb 03, 2016.

November 9th 2015

The call for paper is available.
The submission deadline is January 26, 2016.

October 10th 2015

The website is now available.

Call for Papers

The increased processing capability of mobile and embedded platforms is enabling increasingly ambitious mobile machine vision systems. Industry is actively pursuing embedded vision in the entertainment, automotive and robotics domains. Mobile vision couples high computational requirements with heterogeneous power constrained systems. This makes it an ideal domain in which to evaluate processor architectures, memory efficiency, resource scheduling, mapping, and energy efficient computing. The ASR-MOV workshop brings together system researchers to discuss how the requirements of real-time mobile vision applications impact on tools, architectures and systems..

We invite researchers to contribute both ‘work in progress’ and already published work on a broad range of topics that include, but are not limited to:

  • Domain-specific languages for vision
  • Dynamic and static adaptation techniques
  • Task-based and dataflow models
  • Automatic co-optimization of quality of result and non-functional properties
  • Energy efficiency in heterogeneous SoCs
  • Near-memory and near-camera processing
  • Validation and verification of vision systems
  • Approximate computing in vision
  • Co-design methodologies and case studies
  • Vision processing units (VPUs) compiler-aware techniques


Participants are invited to submit short papers for oral presentation. Accepted paper will be distributed to participants in local proceedings, which will be made available electronically to the participants but will not be published formally. Authors may choose to omit their paper from the local proceedings. Publication at ASR-MOV will not prevent later publication in conferences or journals.


We recommend a 2 to 4 page submission in PDF format, printable on US Letter sized paper. We recommend the template from SIGPLAN:

Submission website:

Please make your submission by the deadline to:

Important Dates:

Submission: Feb 03, 2016
Author Notification: February 10, 2016
Workshop: 12 March 2016.


The workshop starts Saturday 12th of March at 14:00.

14:00 - 14:40 Keynote

Calin Cascaval, Qualcomm Symphony: Orchestrating Heterogeneity for Power Aware Computing
Qualcomm Symphony is a unified framework that integrates performance and power optimizations in a clear and intuitive manner. Symphony enables the deployment of long-running applications, such as video surveillance (recording, encoding, understanding), Advanced Driver Assistance Systems (ADAS), Virtual Reality (VR), and gaming on mobile, embedded platforms. This deployment is attractive because of the low cost, low power, and the flexible form factors of mobile SoCs. However, many of these applications push the power and thermal limits of the platforms, since the platforms are primarily designed to sustain a high level of performance for very short periods of time. Symphony harnesses the power of the platform and provides developers of all skills with an environment in which optimizations for power and thermal are on par with performance optimizations.
About the speaker : Calin Cascaval is Sr. Director in Qualcomm Research Silicon Valley. He is leading the parallel and heterogeneous computing initiatives in Qualcomm. His team developed Qualcomm Symphony, a framework for power aware computing, which is orchestrates execution across different heterogeneous cores on Snapdragon platforms. He also lead the development of the first fully concurrent mobile browser (Zoomm) and other high performance libraries. Previously, he worked at the IBM TJ Watson Research Center on systems software, programming models, and compilers for large scale parallel systems projects, including Blue Gene and PERCS. Calin has a PhD in Computer Science from the University of Illinois at Urbana-Champaign. He has more than 50 peer-reviewed publications and more than 20 awarded patents.

14:40 - 15:30 Session 1

Paul Keir, Porting SLAMBench KFusion to Khronos SYCL.
Amir Momeni, Hamed Tabkhi, Gunar Schirner and David Kaeli, OpenCL-based optimizations for acceleration of object tracking on FPGAs and GPUs.

15:30 - 16:00 break

16:00 - 16:50 Session 2

Harry Wagstaff, Profiling of Vision Applications Using High Speed Simulation.
Julien Martel and Piotr Dudek, Vision Chips with In-pixel Processors for High-performance Low-power Embedded Vision Systems.

17:00 - 17:30 Panel discussion


General Co-chairs

Bruno Bodin, University of Edinburgh, bbodin (A.T)
John Mawer, University of Manchester, john.mawer (A.T)
Luigi Nardi, Imperial College, l.nardi (A.T)

Program Committee

Marco Cornero, ARM, UK
Andrew Davison, Imperial College London, UK
Karol Desnos, INSA Rennes, France
Frank Hannig, Universität Erlangen-Nürnberg, Germany
David Kaeli, Northeastern University, USA
Paul Kelly, Imperial College London, UK
Scott Mahlke, University of Michigan, USA
Richard Membarth, Saarland University, Germany
David Moloney, Movidius, Ireland
Sharathchandra Pankanti, IBM T.J. Watson Research Center, USA