Impact of high level transforms on high level synthesis for motion detection algorithm

High Level Synthesis for System on Chip is a challenging way to cut off development time, while assuming a good level of performance. But the HLS tools are limited by the abstraction level of the description to perform some high level transforms. This paper evaluates the impact of such high level transforms for ASICs and softcores on FPGA. On the representative example of motion detection, we show that we have a speedup of ×1.5 for a softcore on FPGA and ×2.5 for an ASIC while the energy is divided by a factor ×2.90 for the ASIC.

[1]  Alain Trémeau,et al.  A Photometric Model for Specular Highlights and Lighting Changes. Application to Feature Points Tracking. , 2006, 2006 International Conference on Image Processing.

[2]  Jason Helge Anderson,et al.  LegUp: high-level synthesis for FPGA-based processor/accelerator systems , 2011, FPGA '11.

[3]  Michael Fingeroff,et al.  High-Level Synthesis Blue Book , 2010 .

[4]  Jean-Luc Dekeyser,et al.  A High Level Synthesis Flow Using Model Driven Engineering , 2011 .

[5]  Lionel Lacassagne,et al.  Motion detection: Fast and robust algorithms for embedded systems , 2009, 2009 16th IEEE International Conference on Image Processing (ICIP).

[6]  Hunter Scales,et al.  AltiVec Extension to PowerPC Accelerates Media Processing , 2000, IEEE Micro.

[7]  Manuel G. Ortega,et al.  Improved sigma-delta background estimation for vehicle detection , 2009 .

[8]  J. K. Iliffe.  The Use of the Genie System in Numerical Calculation , 1961 .

[9]  E. Martin,et al.  High-level synthesis assisted rapid prototyping for digital signal processing , 2004, Proceedings. The 16th International Conference on Microelectronics, 2004. ICM 2004..

[10]  Minh N. Do,et al.  Youn-Long Steve Lin , 1992 .

[11]  Lionel Lacassagne,et al.  High performance motion detection: some trends toward new embedded architectures for vision systems , 2008, Journal of Real-Time Image Processing.

[12]  Franz Franchetti,et al.  SPIRAL: Code Generation for DSP Transforms , 2005, Proceedings of the IEEE.

[13]  Ken Kennedy,et al.  Optimizing Compilers for Modern Architectures: A Dependence-based Approach , 2001 .

[14]  M. Chinnadurai,et al.  HIGH LEVEL SYNTHESIS , 2011 .

[15]  E.M. Saad,et al.  FPGA-based object-extraction based on multimodal Σ-Δ background estimation , 2009, 2009 2nd International Conference on Computer, Control and Communication.

[16]  Claude Tadonki,et al.  Algorithmic Skeletons within an Embedded Domain Specific Language for the CELL Processor , 2009, 2009 18th International Conference on Parallel Architectures and Compilation Techniques.

[17]  Massimo Piccardi,et al.  Background subtraction techniques: a review , 2004, 2004 IEEE International Conference on Systems, Man and Cybernetics (IEEE Cat. No.04CH37583).