Deforming real-world object using image warping

A system comprising:

• at least one processor;
• and
• a memory component having instructions stored thereon that, when executed by the at least one processor, cause the at least one processor to perform operations comprising:
• applying a machine learning model to a first set of training data comprising a first training image to generate an estimated warping field and estimated segmentation mask;
• computing a deviation between the estimated warping field and the estimated segmentation mask and a ground-truth warping field and segmentation mask corresponding to the first training image in the first set of the training data;
• and
• causing the machine learning model to be applied to a new image to generate a new warping field and segmentation mask to the new image to warp a real-world object depicted in the new image to a target shape.

A non-transitory computer-readable storage medium having stored thereon instructions that, when executed by at least one processor, cause the at least one processor to perform operations comprising:

• applying a machine learning model to a first set of training data comprising a first training image to generate an estimated warping field and estimated segmentation mask;
• computing a deviation between the estimated warping field and the estimated segmentation mask and a ground-truth warping field and segmentation mask corresponding to the first training image in the first set of the training data;
• and
• causing the machine learning model to be applied to a new image to generate a new warping field and segmentation mask to the new image to warp a real-world object depicted in the new image to a target shape.