Back to results
US 2022/0021805 A1Published application
Techniques for providing photographic context assistance
AssigneeSony Corporation
InventorsBrant Candelore, James R. Milne
Life of this patent
Estimating the term…
Abstract
Photographic context assistance is provided by overlaying a composition grid selected by the user or by a machine learning algorithm onto the image of a viewfinder of a camera, and then providing audible and/or visual cues as to moving the aim point to better conform the image to the grid. Fine tuning may be effected using, for instance, stabilization hardware to gently adjust the view to better conform the image to the grid before snapping a shot.
Claims19
A camera comprising:
- at least one display;
- at least one imaging device configured to provide content to the display;
- and
- at least one processor configured with instructions executable to:
- present an image from the imaging device on the display;
- present on the image a composition arrangement;
- and
- present on the camera at least one indication of altering the image using the composition arrangement, wherein the image comprises at least a first object associated with multiple composition arrangements and the composition arrangements for the first object are ranked such that when subsequent objects are recognized in an image, a composition arrangement that best fits the most objects is selected.
5 more elements
An imaging device comprising:
- at least one processor configured with instructions executable to:
- provide photographic context assistance by overlaying a composition grid onto an image of a viewfinder of a camera;
- provide audible and/or visual cues as to moving the camera to better conform the image to the composition grid;
- and
- select the composition grid based on an object in the image identified by a user.
2 more elements
A method for training a machine learning (ML) engine to provide help in obtaining an image using a camera, comprising:
- establishing a training set of object-composition grid pairs;
- inputting the training set to the ML engine;
- and
- providing output of the ML engine processing the training set to an input of the ML engine to further train the ML engine.
1 more element
Dependencies are inferred from the claim text.