These features are, to some extent, invariant to illumination,
#Cartoon picture shapes Patch
Local patch based feature have been proposed and successfully applied in objectĭetection, such as Scale-Invariant Feature Transform (SIFT) ( Lowe,ġ999), Speeded Up Robust Features (SURF) ( Bay et al.,Ģ008), Maximally Stable Extremal Regions (MSER) ( MatasĮt al., 2002) and scale and affine invariant interest point ( MikolajczykĪnd Schmid, 2004).
The character usually localize in part of the whole picture ( Szeliski, Global features usually contain too much noise because Matches between object features and the image features ( MikolajczykĪnd Schmid, 2005). The central idea of the general object detection problem lies in finding Is no previous proper method for detection the characters in the 2-D cartoon The chief value of cartoons is the characters in them and the characters areĪlso the most frequently pirated content. The sharing web sit to detect and then reject the cartoons with copyright statement. Legal liability for spreading pirate cartoons. Way to the sharing website, so that the sharing website can be protected from Some pirates sometimes even re-edit the cartoon for advertisement, create newĬartoon product based on the famous cartoon characters and publish as theirĪ method to solve the pirate problem of cartoon is to block pirates May get and upload the cartoon media copy to sharing website to make a profit. The economic value of cartoon attracts both the cartoon fans and pirate, who Had been produced by artists and shown to readers which formed a big industry. In history, lots of classic cartoon characters Information Technology Journal, 12: 2342-2349.Ĭartoon is well-liked by both children and adults for its comic charactersĪnd the funny drawing style. 2-D Cartoon Character Detection based on Scalable-Shape Context and Hough Voting. Abd El-Latif, Xuefeng Bai and Xiamu Niu, 2013. The experimental results show that the proposed SSC-based detection method is effective in the detection of 2D-cartoon characters. Finally, a Hough-voting scheme is employed to find the location of the character in the testing image. Then, the matching problem between the key points extracted from the input model and testing image is solved as an optimal assignment problem. Secondly, the scale of each key point is used as a reference scale for Shape Context (SC) to describe the curvilinear structure around the key points. Firstly, we use the Harris-Laplace corner detector to find the key points at multi-scale in the cartoon image, most of which are localized at the junctions of curves. We extract the curve in the cartoon image as the main content and then design a local shape feature named Scalable-Shape Context (SSC) to present the local shape of cartoon. In this study, we propose a new method to detect the characters in 2D-cartoon images, aiming at rejecting pirate uploading automatically. Cartoon pirate uploading is a very serious problem for the image and video-sharing website.
0 kommentar(er)
