US 2018/0005077 A1 · AUTOMATED SELECTION OF SUBJECTIVELY BEST IMAGE FRAMES FROM BURST CAPTURED IMAGE SEQUENCES

Abstract

A “Best of Burst Selector,” or “BoB Selector,” automatically selects a subjectively best image from a single set of images of a scene captured in a burst or continuous capture mode, captured as a video sequence, or captured as multiple images of the scene over any arbitrary period of time and any arbitrary timing between images. This set of images is referred to as a burst set. Selection of the subjectively best image is achieved in real-time by applying a machine-learned model to the burst set. The machine-learned model of the BoB Selector is trained to select one or more subjectively best images from the burst set in a way that closely emulates human selection based on subjective subtleties of human preferences. Images automatically selected by the BoB Selector are presented to a user or saved for further processing.

Claims20

Dependent claims

A computer-implemented process, comprising:

receiving an arbitrary burst set comprising a plurality of images of a scene captured via an imaging device;
receiving a machine-learned predictive best of burst model;
the machine-learned predictive best of burst model automatically generated from a combination of features extracted from a plurality of training burst sets, each training burst set comprising a plurality of images of a scene, and corresponding human selections of a single subjective favorite image from each training burst set;
applying the machine-learned predictive best of burst model to rank each of the images in the arbitrary burst set in terms of a predicted subjective goodness level;
and
outputting a highest ranked one of the images of the arbitrary burst set as a predicted subjectively best image.

3 more elements

A system, comprising:

a general-purpose computing device;
and
a computer program comprising program modules executable by the computing device, wherein the computing device is directed by the program modules of the computer program to:
receive a machine-learned predictive model;
the predictive model automatically trained on a combination of image features extracted from a plurality of training burst sets and corresponding human selections of favorite images from each training burst set;
each training burst set comprising a plurality of images of a scene;
receive an arbitrary burst set comprising a plurality of images of a scene;
extract a plurality of features from each image of the arbitrary burst set;
apply the predictive model to the features extracted from the images of the arbitrary burst set to predict a subjective goodness level for each of the images in the arbitrary burst set;
and
output one or more of the images in the arbitrary burst set having the highest predicted subjective goodness levels relative to the other images in the arbitrary burst set.

8 more elements

A computer-readable storage device having computer-executable instructions stored thereupon which, when executed by a computer, cause the computer to:

receive a machine-learned predictive best of burst model, the predictive best of burst model automatically trained on a combination of features extracted from a plurality of training burst sets and corresponding human selections of a single subjective favorite image from each training burst set;
each training burst set comprising a plurality of images of a scene captured via an imaging device;
receive an arbitrary burst set comprising a plurality of images of a scene captured via an imaging device;
apply the machine-learned predictive best of burst model to rank each of the images in the arbitrary burst set in terms of a predicted subjective goodness level;
and
output a highest ranked one of the images in the burst set as a predicted subjectively best image.

3 more elements

Dependencies are inferred from the claim text.

Automated selection of subjectively best image frames from burst captured image sequences

Life of this patent

Automated selection of subjectively best image frames from burst captured image sequences

Life of this patent

Abstract

Claims20