Imaging workflow using facial and non-facial features
A method for determining an impact score for a digital image includes providing the digital image wherein the digital image includes faces; using a processor to determine an image feature for the faces; using the processor to compute an object impact score for the faces, wherein the object impact score is based at least upon one of the determined image features; weighting the object impact score for the faces based on one of the determined image features for a face; using the processor to compute an impact score for the digital image by combining the weighted object impact scores for the faces in the image; and storing the computed impact score in a processor accessible memory.
1. A method, implemented as computer software and executed via a computer processor, for determining an aesthetic quality parameters of a collection media assets, the method comprising:
for each media asset in the collection of media assets:
accessing a media asset from the collection of media assets;
detecting objects appearing in the media asset, wherein the objects comprise people;
extracting object features from the objects in the media asset, wherein the object features comprise facial features;
detecting an asset feature in the media asset, wherein the asset feature is selected from the group consisting of: brightness, colorfulness, number of hues, color distribution, sharpness, noise, and contrast;
detecting a capture feature, wherein the capture feature comprises information about the media asset obtained at the capture of the media asset;
detecting a social feature of the media asset, wherein the social feature comprises information about the media asset obtained via social interaction;
computing an aesthetic quality parameter based on the object features, the asset feature, the capture feature, and the social feature, wherein the aesthetic quality parameter is a multidimensional vector comprising disparate types of values;
associating the aesthetic quality parameter with the media asset; and
identifying which media asset in the collection of media assets has the highest aesthetic quality based on the aesthetic quality parameter assigned to each media asset in the collection of media assets.
2. The method of claim 1 wherein detecting objects appearing in the media asset comprises detecting human faces.
3. The method of claim 1 wherein the facial features are selected from the group consisting of: face size, face location, face pitch, face roll, face yaw, face contrast, face brightness, face color balance, face noise, gender, race, hair style, neatness of appearance, identity, facial expression, eye blink, eye gaze, red eye, mouth open, teeth detection, tongue detection, glasses detection, hat detection, and occlusion detection of the face.
4. The method of claim 1 wherein the object features further comprises clothing features, wherein the clothing features are selected from the group consisting of: clothing color, clothing style, clothing condition.
5. The method of claim 1 wherein the object features further comprises environmental features.
6. The method of claim 1 wherein the objects further comprise human bodies, human torsos, animals, bacterium, viruses, internal organs, cars, and assembly line parts.
7. The method of claim 1 wherein the capture feature comprises one or more of the following types of information: date and time that the media asset was captured, focal length, subject distance, magnification, presence of a flash, presence of a self-timer, GPS coordinates, and image resolution.
8. The method of claim 1 wherein the social feature comprises one or more of the following types of information: number of likes of the media asset, comments, tags, viewings, downloads, links to other photo collections, upload date, and whether the media asset was used as a profile picture.
9. The method of claim 1 wherein computing the aesthetic quality parameter comprises summing up equally weighted inputs of the object features, the asset feature, the capture feature, and the social feature.
10. The method of claim 1 wherein detecting the asset feature in the media asset comprises detecting brightness, colorfulness, number of hues, color distribution, sharpness, noise, and contrast for each of the objects appearing in the media asset.