Abstract:With the development of Internet and cloud computing, multimedia data, such as images and videos, has become one of the most common data types being processed. As the scale of multimedia data being still increasing, it is vitally important to efficiently extract useful information from such a huge amount of multimedia data. However, due to the complexity of the core algorithms, multimedia retrieval applications are not only data intensive but also computationally intensive. Therefore, it has been a major challenge to accelerate the processing speed of such applications to satisfy the real-time requirement. As Graphic Processing Unit (GPU) has entered the general-propose computing domain (GPGPU), it has become one of the most popular accelerators for the applications with real-time requirements. In this paper, we parallelize a widely-used image retrieval algorithm called SURF on GPGPU, which is the core algorithm for many video and image retrieval applications. We first analyze the parallelism within SURF to guarantee that there are sufficient tasks being mapped to the large-scale computation resources in GPGPU. We then exploit some inherent GPGPU characteristics, such as 2D memory, to further boost the performance. Finally, we provide some optimization to the cooperation between CPU and GPGPU, which is generally ignored in previous designs. Experimental results show that our parallelization and optimization achieve a throughput of 340.5 frames/s on a NVIDIA GTX295 GPGPU, which is 15X faster than the maximal optimized CPU version. Compared to CUDA SURF, a state-of-the-art parallelization of SURF on GPGPU, our system achieves a speedup by a factor of 2.3X.

Gpu Based Fast Algorithm For Tanner Graph Based Image Interpolation

Matrix Decomposition Graphics Processing Unit Solver for Poisson Image Editing

Global optimization based image inpainting andits implementation on GPU

GPU-friendly shape interpolation based on trajectory warping

GPU-Based Texture Encoding and Real-Time Rendering

A Practical Algorithm For Tanner Graph Based Image Interpolation

Technical Section: Real-time B-spline Free-Form Deformation via GPU acceleration

GPU Accelerated Computation for Surface Topography Measurement

Real-Time Surface Topography Measurement with Graphic Processing Unit

Real-time B-spline Free-Form Deformation via GPU acceleration.

PAR Model SAR Image Interpolation Algorithm on GPU with CUDA

Parallel Motion Compensation Interpolation in H.264/AVC Using Graphic Processing Units

Balancing Cpu And Gpu: Real-Time Visualization Of Large Scale 3d Scanning Models

Efficient Gpu-Based Inter Prediction For Video Decoder

A GPU-based high-throughput image retrieval algorithm

Fast Shape-Simplifying Image Abstraction Using Graphics Hardware.

Research on the fast Fourier transform of image based on GPU

Accelerated 3-D T-FDK Reconstruction Algorithm Using the Commordity Graphics Hardware

Parallel Computing Method of Commonly Used Interpolation Algorithms for Remote Sensing Images

Rapid Evaluation of B-spline Surfaces Using GPU

The Accelerated Fusion Algorithm of Remote Sensing Image Data Based on GPU