基于GPU的HDR图像平移配准加速算法
Accelerating Translational Image Registration for HDR Images on GPU
摘要：
高动态范围（HDR）图像是使用场景的多次曝光生成的。当使用手持相机捕捉静态场景时，需要通过在两个维度上全局移动每个图像来对齐这些图像。为了实现快速和稳健的对齐，通常使用中值阈值位图（MTB）和创建图像金字塔来计算移位量。在这项研究中，我们优化这些计算使用并行处理的方法，利用GPU。实验结果表明，该实现实现的速度高达6.24倍的基线多线程CPU实现的对齐的一个图像对。源代码可在https://github.com/kadircenk/WardMTBCuda上获得
英文摘要：
High Dynamic Range (HDR) images are generated using multiple exposures of a scene. When a hand-held camera is used to capture a static scene, these images need to be aligned by globally shifting each image in both dimensions. For a fast and robust alignment, the shift amount is commonly calculated using Median Threshold Bitmaps (MTB) and creating an image pyramid. In this study, we optimize these computations using a parallel processing approach utilizing GPU. Experimental evaluation shows that the proposed implementation achieves a speed-up of up to 6.24 times over the baseline multi-threaded CPU implementation on the alignment of one image pair. The source code is available at https://github.com/kadircenk/WardMTBCuda
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
