Interaction method for regions of-interest in time series images

A method for identifying a region of interest within a time sequence of images, comprising:

• acquiring a time sequence of images comprising a plurality of image frames;
• performing image segmentation to segment a first region of interest (ROI) and a second ROI from within each of the plurality of image frames of the time sequence of images;
• receiving manual edits to the first ROI within a first edited image frame for the first ROI of the plurality of image frames;
• receiving manual edits to the first ROI within a second edited image frame for the first ROI of the plurality of image frames;
• receiving manual edits to the second ROI within a first edited image frame for the first ROI of the plurality of image frames;
• receiving manual edits to the second ROI within a second edited image frame for the first ROI of the plurality of image frames;
• propagating the manual edits to the first ROI within the first edited image frame for the first ROI to image frames of the plurality of images preceding the first edited image frame for the first ROI and to image frames between the first edited image frame for the first ROI and the second edited image frame for the first ROI, to an extent that is dependent upon a first transformation function used to propagate the manual edits to the first ROI within the first edited image and a first weighing factor that is dependent upon a distance between the image frames and the first edited image frame for the first ROI;
• propagating the manual edits to the first ROI within the second edited image frame for the first ROI to image frames of the plurality of images subsequent to the second edited image frame for the first ROI and to image frames between the first edited image frame for the first ROI and the second edited image frame for the first ROI, to an extent that is dependent upon a second transformation function used to propagate the manual edits to the first ROI within the second edited image for the first ROI and a second weighing factor that is dependent upon a distance between the image frames and the second edited image frame for the first ROI;
• propagating the manual edits to the second ROI within the first edited image frame for the second ROI to image frames of the plurality of images preceding the first edited image frame for the second ROI and to image frames between the first edited image frame for the second ROI and the second edited image frame for the second ROI, to an extent that is dependent upon a third transformation function used to propagate the manual edits to the second ROI within the first edited image and a first weighing factor that is dependent upon a distance between the image frames and the first edited image frame for the second ROI;
• and
• propagating the manual edits to the second ROI within the second edited image frame for the second ROI to image frames of the plurality of images subsequent to the second edited image frame for the second ROI and to image frames between the first edited image frame for the second ROI and the second edited image frame for the second ROI, to an extent that is dependent upon a fourth transformation function used to propagate the manual edits to the first ROI within the second edited image for the second ROI and a second weighing factor that is dependent upon a distance between the image frames and the second edited image frame for the second ROI,
• wherein the image frames between the first and second edited image frames for the first ROI transition from being fully influenced by the manual edits of the first edited image frame for the first ROI and not at all influenced by the manual edits of the second edited image frame to being fully influenced by the manual edits of the second edited image frame for the first ROI and not at all influenced by the manual edits of the first edited image frame for the first ROI, and
• wherein the image frames between the first and second edited image frames for the second ROI transition from being fully influenced by the manual edits of the first edited image frame for the second ROI and not at all influenced by the manual edits of the second edited image frame to being fully influenced by the manual edits of the second edited image frame for the second ROI and not at all influenced by the manual edits of the first edited image frame for the second ROI.

A computer system comprising:

• a processor;
• and
• a non-transitory, tangible, program storage medium, readable by the computer system, embodying a program of instructions executable by the processor to perform method steps for identifying a region of interest within a time sequence of images, the method comprising:
• acquiring a time sequence of images comprising a plurality of image frames;
• performing image segmentation to segment a first region of interest (ROI) and a second ROI from within each of the plurality of image frames of the time sequence of images;
• receiving one or more manual edits to the first ROI within a first keyframe for the first ROI of the plurality of image frames;
• receiving one or more manual edits to the ROI within a second keyframe for the first ROI of the plurality of image frames;
• receiving one or more manual edits to the second ROI within a first keyframe for the second ROI of the plurality of image frames;
• receiving one or more manual edits to the second ROI within a second keyframe for the second ROI of the plurality of image frames;
• propagating the manual edits corresponding to the first ROI to other image frames of the plurality of images, the other image frames being those image frames of the plurality of image frames that are neither the first keyframe for the first ROI nor the second keyframe for the first ROI;
• and
• propagating the manual edits corresponding to the second ROI to other image frames of the plurality of images, the other image frames being those image frames of the plurality of image frames that are neither the first keyframe for the second ROI nor the second keyframe for the first ROI,
• wherein an extent to which each of the manual edits for the first and second ROI are propagated to other image frames is dependent upon a distance in time between each of the other image frames and the first and second keyframes that have been manually edited for the first ROI, respectively, and
• wherein the first keyframe for the first ROI, the second keyframe for the first ROI, the first keyframe for the second ROI, and the second keyframe for the second ROI are all different frames of the plurality of image frames.