A computing system and method that can be used for performing transformations of images using machine-learned model(s) and interpretable user-specified parameters that control the transformations. More specifically, the machine-learned model can be used to perform transformation(s) on an image based on different parameter(s) of a user-specified conditioning vector. The systems and methods of the present disclosure allow for a user to specify desired values for defined characteristics of a transformation (e.g., area of transformation, degree of transformation, etc.) for one or more transformations of the defined characteristic(s) (e.g., transforming lack of facial displacement distortion to presence of a facial displacement distortion, transforming glasses frames to lack of glasses frames, unshaded to shaded, etc.).
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Ramos, Igor and Klein, Daniel, "Image and/or Video Transformation Using Interpretable Transformation Parameters", Technical Disclosure Commons, (November 15, 2021)