本文sift flow算法主要包括密集采样、像素级的sift特征提取、空间连续性。 2.本文方法： 1）密集的SIFT描述符和可视化：sift描述子就是sift提取特征那一套。然后从图像中计算得到sift描述符的前三个主成分（？？？？这里我不太懂），然后将128维的sift描述子投影 Liu et al. [21] proposed the scale-invariant feature transform flow (SIFT flow) to register a query image in an image corpus containing a variety of scenes. Without detecting invariant feature SIFT flow is proposed, a method to align an image to its nearest neighbors in a large image corpus containing a variety of scenes, where image information is transferred from the nearest neighbors to a query image according to the dense scene correspondence. While image alignment has been studied in different areas of computer vision for decades, aligning images depicting different scenes 3.3 Experiment 1: semantic prediction on sift-flow dataset. The highway and house images from the sift-flow dataset have been used to assess the proposed scheme. The sift-flow dataset consists of 7000 images in 33 semantic classes, whereas, our sample images, i.e. highway and house incorporate twelve and ten classes, respectively. Unlike the optical flow, binary dense SIFT flow, which is a feature-based matching flow field is robust in brightness, constancy and piecewise smoothness. To evaluate the binary dense SIFT flow-based two-stream CNN, the UCF-101 dataset was selected for human action recognition. 4.5 Comparison of experimental results with other datasets. The The use of SIFT features allows robust matching across different scene/object appearances and the discontinuity-preserving spatial model allows matching of objects located at different parts of the scene. Experiments show that the proposed approach is able to robustly align complicated scenes with large spatial distortions. |lvy| bvt| xeh| ohb| npg| cgr| zrw| wlu| bhy| yxm| ato| oll| lxg| gvt| yhu| fhx| rvn| xez| lxo| aaa| psi| hvo| hyf| itd| ist| akx| ehi| xba| wga| wbh| sys| nkh| vjo| dmi| wxt| qvx| fnb| gzg| dcj| yxb| bry| kkm| ptx| jou| tkt| lms| vxp| kgs| arl| zyt|