首页|Researchers Submit Patent Application, 'Generating Masked Regions Of An Image Us ing A Predicted User Intent', for Approval (USPTO 20240201832)
Researchers Submit Patent Application, 'Generating Masked Regions Of An Image Us ing A Predicted User Intent', for Approval (USPTO 20240201832)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
News editors obtained the following quote from the background information suppli ed by the inventors:“Digital tools allow artists to interact with objects (and sub-objects) in an image. To indicate an objectto be interacted with, users use one of multiple tools to select an object of interest. The tool selected andus ed by the user to indicate the object of interest is based on the user’s accurac y indicating the object ofinterest, the user’s gesture to indicate the object o f interest, and the like. For example, a user may selectone tool of multiple to ols based on the gesture the user intends to perform (selecting, lassoing, traci ng,boxing, stroking, etc.). Tools are used improperly when the user uses the wr ong gesture with the wrongtool. For example, a user may use a bounding box tool but indicate the object of interest using a lasso.Because the bounding box too l does not expect a lasso gesture, the tool fails to correctly capture theobjec t of interest. When the appropriate tool is used with the appropriate gesture (e .g., a selection toolreceives a user selection of an object of interest), the o bject of interest is captured. After the object ofinterest is captured, the use r may user may adjust attributes of the object (e.g., size, color, shape, opacity, position), add similar objects, delete the object, and the like.”
NETL
NSTL
