Implement the Fourier Correlation Algorithm

Fourier correlation has many applications:

• Measure the similarity of two signals
• Find the best translation to overlay similar images
• Segment volumetric medical images

A Fourier correlation algorithm† can be composed from Intel® oneAPI Math Kernel Library (oneMKL) functions for easy offload to various accelerators such as GPUs and FPGAs.

In this session, senior principal software engineer Henry Gabb:

• Covers the basics of:
  • Data Parallel C++ (DPC++) queues
  • Buffers, Unified Shared Memory (USM), and data movement
  • Implicit vs. explicit synchronization
• Walks through buffered and USM implementations of a 1D Fourier correlation
• Discusses the movement of data between host and device and ways to avoid unnecessary data movement

† correlation=MAXLOC(IDFT(signal1)*CONJG(DFT(SIGNAL2)))) where DFT is the discrete Fourier transform, IDFT is the inverse DFT, CONJG is the complex conjugate, and MAXLOC is the location of the maximum correlation score.

Download the On-Demand Video

Download the Presentation

• Get DPC++ and oneMKL as part of the Intel® oneAPI Base Toolkit—a foundational set of tools and libraries for developing high-performance, data-centric applications across diverse architectures.
• Download the stand-alone version of oneMKL.

• Learn more about DPC++.
• Sign up for an Intel® Developer Cloud account—a free development sandbox with access to the latest Intel hardware and oneAPI software.
• Explore oneAPI including developer opportunities and benefits.
• Subscribe to Code Together, an interview series that explores the challenges at the forefront of cross-architecture development. Each biweekly episode features industry VIPs who are blazing new trails through today's data-centric world. Available wherever you get your podcasts.