Machine guided photo and video composition

1. A method comprising:

generating processed image data with reduced pixel channel values from a raw image data through operation of an image processor;

operating an object detection engine to identify objects and object locations in the processed image data through use of an object detection algorithm;

operating a computer vision analysis engine to identify geometric attributes of objects in the processed image data through use of a line geometry analysis algorithm;

operating an image cropping engine to:

select potential cropped image locations within the processed image data through operation of a selector;

generate crop location scores for each of the potential cropped image locations and determine highest scored cropped image location through operating of a cropped image evaluator; and

communicate highest crop location score for the highest scored cropped image location to a score evaluator gate;

generating object classifications from the object locations and the geometric attributes through operation of an object classifier;

receiving device instructions at a user interface controller by way of the score evaluator gate; and

configuring a display device to display device positioning instructions transformed from the received device instructions.

2. The method of claim 1 comprising:

receiving the object locations and the geometric attributes at a switch configured by an avoid object amputation control to release the object locations and the geometric attributes to the image cropping engine, wherein the object locations and the geometric attributes identify the object locations as bounding boxes around the objects; and

operating the image cropping engine to:

select the potential cropped image locations within the processed image data from the object locations and the geometric attributes through operating of a selector, wherein the selector excludes the potential cropped image locations with partial bounding boxes around the objects.

3. The method of claim 1 , wherein the object detection algorithm utilizes a convolutional neural network machine learning algorithm to identify the objects and the object locations in the processed image data.

4. The method of claim 1 , wherein the cropped image evaluator utilizes a convolutional neural network machine learning algorithm to determine the highest scored cropped image location.

5. The method of claim 1 further comprising:

releasing a control from the score evaluator gate to a crop coordinate mapper in response to the highest crop location score being greater than a crop score threshold;

generating the device instructions comprising device zoom, focus, and rotation instructions from the crop coordinate mapper in response to receiving the control from the score evaluator gate; and

communicating the device instructions to the user interface controller from the crop coordinate mapper.

6. The method of claim 1 further comprising:

releasing a control from the score evaluator gate to an object based compute engine in response to the highest crop location score being lower than a crop score threshold;

generating the device instructions comprising device focus and rotation instructions from the object locations, the geometric attributes, and the object classifications through operation of the object based compute engine in response to receiving the control from the score evaluator gate; and

communicating the device instructions to the user interface controller from the object based compute engine.

7. The method of claim 1 , wherein the user interface controller generates revised device positioning instructions in response to a device position information from the display device.