This publication describes techniques for implementing virtual camera image transformations. It can be difficult to properly position cameras to capture certain types of photographs, particularly pictures intended to include the photographer (often referred to as “selfies”). Even with the use of assistive devices, such as “selfie sticks,” these types of photographs often exhibit a characteristic, unnatural look due to suboptimal camera orientation and/or focal length (since the relative distance from the lens and the field of view of that lens affect the look of the subjects within the image). Disclosed in this publication is a virtual camera image processor capable of adjusting, inter alia, camera angle, position, and/or focal length of images during post-processing (e.g., as an after effect) using a machine-learned image generator.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.