Image synthesis with multiple input modalities

A method, comprising:

• displaying a user interface on at least one display device, wherein the user interface comprises:
• at least one input area operable to display graphical indications that represent one or more user generated inputs, wherein each user generated input corresponds to one of two or more distinct input modalities and is generated with one or more selectable input generation tools;
• an output area operable to display an output image generated based on a utilized subset of the one or more user generated inputs;
• and
• at least one input utilization control, distinct from the input generation tools, comprising a first displayed set of input modality descriptors and operable to enable a user to determine the utilized subset by selecting any combination of the input modality descriptors from the first displayed set and thereby to indicate which ones of the generated user inputs should influence the output image;
• wherein the input modalities comprise:
• a segmentation input modality in which the input area may be partitioned into one or more segments;
• and
• a sketch input modality in which lines or curves may be drawn in the input area;
• wherein the lines or curves drawn with the sketch input modality need not partition the input area, such that the sketch input modality and the segmentation input modality are distinct input modalities.

A non-transitory computer-readable medium having instructions stored therein that, if executed by one or more processors in one or more computing devices, cause the computing devices to perform steps comprising:

• displaying a user interface on one or more display devices, wherein the user interface comprises:
• two or more input generation tools, each corresponding to one of a plurality of distinct input modalities and each operable to generate a user input having an associated user input graphical indication;
• at least one input area operable to display the user input graphical indications associated with a visualized subset of generated user inputs;
• an output area distinct from the input area and operable to display an output image that is based on a utilized subset of the generated user inputs;
• at least one input utilization control, distinct from the input generation tools, comprising a displayed collection of input modality descriptors and operable to enable a user to determine the utilized subset by selecting one or more of the input modality descriptors in the displayed collection of input modality descriptors and thereby to indicate which ones of the generated user inputs should influence the output image;
• and
• at least one input visualization control distinct from the input generation tools and distinct from the input utilization control and operable to determine the visualized subset by selecting which ones of the generated user inputs should have their graphical indications displayed in the input area;
• wherein the selection of which ones of the generated user input graphical indications should be displayed in the input area is independent of the selection of which ones of the generated user inputs should influence the output image.