Acceleration of High Dynamic Range Imaging Pipeline Based on Multi-threading and SIMD Technologies
نویسندگان
چکیده
In this paper we present a holistic approach to CPU based acceleration of the high dynamic range imaging (HDRI) pipeline. The high dynamic range representation can encode images regardless of the technology used to create and display them, with the accuracy that is only constrained by the limitations of the human eye and not a particular output medium. Unfortunately, the increase in accuracy causes significant computational overhead and effective hardware acceleration is needed to ensure a utility value of HDRI applications. In this work we propose a novel architecture of the HDRI pipeline based on CPU SIMD and multi-threading technologies. We discuss the impact on processing speed caused by vectorization and parallelization of individual image processing operations. A commercial application of the new HDRI pipeline is described together with evaluation of achieved image processing speedup.
منابع مشابه
General-Purpose Computation Using Graphics Hardware for Fast HDR Image Processing
This paper presents a new approach to hardware aided image processing and analysis, primarily focused on HDR imaging. In order to achieve interactive frame rates and great processing speeds we propose a library model which is able to utilize efficiently most powerful of the underlying vector hardware. The resulting library architecture was implemented and tested on GPUs and SIMD capable multico...
متن کاملSynergy: A HW/SW Framework for High Throughput CNNs on Embedded Heterogeneous SoC
Convolutional Neural Networks (CNN) have been widely deployed in diverse application domains. There has been significant progress in accelerating both their training and inference using high-performance GPUs, FPGAs, and custom ASICs for datacenter-scale environments. The recent proliferation of mobile and IoT devices have necessitated real-time, energy-efficient deep neural network inference on...
متن کاملDynamic Warp Subdivision for Integrated Branch and Memory Divergence Tolerance : Extended Tradeoff Analysis ∗
SIMD organizations amortize the area and power of fetch, decode, and issue logic across multiple processing units in order to maximize throughput for a given area and power budget. However, throughput is reduced when a set of threads operating in lockstep (a warp) are stalled due to long latency memory accesses. The resulting idle cycles are extremely costly. Multi-threading can hide latencies ...
متن کاملHigh Dynamic Range Imaging
High dynamic range (HDR) images and video contain pixels, which can represent much greater range of colors and brightness levels than that offered by existing, standard dynamic range images. Such “better pixels” greatly improve the overall quality of visual content, making it appear much more realistic and appealing to the audience. HDR is one of the key technologies of the future imaging pipel...
متن کاملImplementation of HDR Photographic Pipeline in Mobile Devices
The high dynamic range (HDR) photography offers significantly better quality than conventional digital images stored in low dynamic range (LDR) formats. We replace the standard image capture pipeline in mobile phones with an HDR acquisition pipeline based on the multi-exposure method. We discuss limitations of the mobile platforms that affect architecture of the proposed pipeline, then implemen...
متن کامل