A framework for optimizing OpenVX applications performance on embedded manycore accelerators

Abstract

Nowadays Computer Vision application are ubiquitous, and their presence on embedded devices is more and more widespread. Heterogeneous embedded systems featuring a clustered manycore accelerator are a very promising target to execute embedded vision algorithms, but the code optimization for these platforms is a challenging task. Moreover, designers really need support tools that are both fast and accurate. In this work we introduce ADRENALINE, an environment for development and optimization of OpenVX applications targeting manycore accelerators. ADRENALINE consists of a custom OpenVX run-time and a virtual platform, and overall it is intended to provide support to enhance performance of embedded vision applications.

Software defined multi-spectral imaging for Arctic sensor networks

“…This hypothesis that benchmarks can assist with architectural computing decisions can only be confirmed by expanding the benchmark suite and completion and testing of applications built from them. Similar work to identify common transforms for direct acceleration in co-processors is in progress by the OpenVX standards effort [32]. Ultimately, it is likely that a full suite of transform primitives for computer and machine vision can be identified so that image transform acceleration becomes much like graphics today, where all fundamental transforms are handled by custom application specific integrated circuit solution co-processors.…”