The final blog in this series describes efforts to 1) extract rectilinear images (field of view) from equirectangular images (360-degree panoramic images) utilizing a GPU and 2) execute the code on both the CPU and GPU simultaneously. Since the previously discussed open-source code is available at https://github.com/intel-health/optimizing-equirectangular-conversion and it utilizes oneAPI constructs, executing on other hardware options becomes straightforward.
The last blog, GPU Exploration, provides information about the performance of the code on the GPU and discusses code changes necessary to adapt the code to allow both the CPU and GPU to process the data in parallel.
The full series describes a developer journey that begins with learning about new technology, finding available open-source implementations, determining methods to optimize the execution speed with serial techniques, then parallel techniques, and finally culminates with how multiple pieces of hardware can be used simultaneously. Tools such as Intel® VTune, Intel Advisor, and Intel oneAPI Base Toolkit are highlighted for their utility in identifying and implementing potential optimizations.
- Introduction to 360-Degree Videos and Imaging
- Existing Open Source for 360-Degree Projection
- Execution Framework and Serial Code Optimizations
- Execution Framework and Parallel Code Optimizations
- GPU Exploration
Let us know in the comments if there are any questions about the information found in these blog postings.
If you are interested in being notified about other health and life science postings, subscribe to get the latest updates from the team.
About the Intel Health and Life Sciences
Contact Intel's Health and Life Sciences team at health.lifesciences@intel.com or learn more about what we do at https://www.intel.com/health.
About the Author
Doug Bogia received his Ph.D. in computer science from the University of Illinois, Urbana-Champaign, and works at Intel Corporation. He enjoys photography, woodworking, programming, and optimizing solutions to run as fast as possible on a given piece of hardware. Connect with him on LinkedIn https://www.linkedin.com/in/doug-bogia-4653696/ by mentioning this blog.
