Scale Invariance Feature Transform (SIFT) is quite suitable for image matching because of its invariance to image scaling, rotation and slight changes in illumination or viewpoint. However, due to high computation com...
详细信息
Scale Invariance Feature Transform (SIFT) is quite suitable for image matching because of its invariance to image scaling, rotation and slight changes in illumination or viewpoint. However, due to high computation complexity it's technically challenging to deploy SIFT in real time application situations. To address this problem, we propose CLSIFT, an OpenCL based highly speeded up and performance portable SIFT solution. Important optimization techniques employed in CLSIFT such as: (1) For less global memory traffic, independent logical functions are merged into the same kernel to reuse data. (2) loop buffers are introduced in for data and intermediate results reusing. (3)Task queue used to schedule threads in the same branch to remove branch divergences. (4) Data partition is based on the statics patterns for workload balance among workgroups. (5) Overlap of CPU time and better parallel strategies are used too. With all mentioned efforts, CLSIFT processes lena. jpg at 74.2 FPS and 43.4FPS respectively on NVidia and AMD GPUS, much higher than CPU's nearly 10 FPS and the known fastest SIFTGPU's 39.8 FPS and 13FPS. Moreover in a quantitative comparison approach we analyze those successful strategies beating SIFTGPU, a famous existing GPU implementation. Additionally, we observe and conclude that NVidia GPU achieves better occupancy and performance due to some factors. Finally, we summarize some techniques and empirical guiding principles that may be shared by other applications on GPU.
We introduce a model considering the single-leg revenue management problem from the perspective of online algorithms and competitive analysis. In this model, the price and limitation of bookings are both decision vari...
详细信息
A novel multichannel multiple access (MMA) protocol is proposed for vehicular ad hoc networks in this paper. For increasing channel efficiency, two control channels are utilized in MMA to finish request-to-send/clear-...
详细信息
To enhance classification performance by making use of easily availab.e unlab.lled data to overcome the scarcity of lab.lled data, this paper proposes an Embedded Co-Adaboost algorithm that integrates multi-view learn...
详细信息
A two-level distributed state estimation method for substation-dispatch center is proposed. After zero impedance three-phase state estimation based on KCL has been applied in substation, up load ripe data are treated ...
详细信息
In view of using only the intensity information of retroreflection wave and poor capacity of anti-interference in laser short-range detection,this paper analyzes the polarization characteristics of laser beam on the m...
详细信息
ISBN:
(纸本)9781627483766
In view of using only the intensity information of retroreflection wave and poor capacity of anti-interference in laser short-range detection,this paper analyzes the polarization characteristics of laser beam on the metal surface and establishes reflection model of polarized laser.A target detection method based on polarization characteristics of reflection laser is putted forward and the simulation of polarization characteristics on the surface of several typical metal targets is implemented at last.
PM marker along with an electrical system sensing its magnetic field is commonly used to determine the parts locations in industry. This paper proposes a novel method combining computation and experiment to improve th...
详细信息
During long-time operation, thermal deformation of heat exchange pipe weaken the capability and reduce the reliability of heat transfer system, especially the heat exchange pipe with initial bending in high temperatur...
详细信息
Thermal deformation of heat exchange pipe is the significant source of structural variation in ground heat exchange system, which could weaken the capability and reduce the reliability of heat transfer system after lo...
详细信息
This paper is concerned with identification of linear parameter varying (LPV) systems in an input-output setting with Box-Jenkins (BJ) model structure. Classical linear time invariant prediction error method (PEM) is ...
详细信息
暂无评论