Techniques for user customization in a photo management system

A computer-implemented method comprising:

• receiving, at a computing device including one or more processors, a plurality of photos;
• selecting, at the computing device, a subset of the plurality of photos by performing joint global maximization of photo quality and photo diversity using a quality metric for each of the plurality of photos and a similarity matrix for the plurality of photos, the quality metric for a specific photo being indicative of a quality of the specific photo, the similarity matrix being indicative of a similarity between each unique pair of photos of the plurality of photos;
• determining, at the computing device, one or more clusters for the subset of the plurality of photos using a hierarchical clustering algorithm, the one or more clusters collectively including less than or equal to all of the subset of the plurality of photos, each of the one or more clusters including one or more distinct photos from the subset of the plurality of photos;
• storing, at the computing device, the subset of the plurality of photos arranged based on the one or more clusters;
• receiving, at the computing device, user input with respect to the subset of the plurality of photos arranged based on the one or more clusters, the user input corresponding to at least one of adding a first specific photo from a remainder of the plurality of photos to a specific cluster and deleting a second specific photo from the specific cluster;
• when the user input corresponds to at least one of adding the first specific photo from the remainder of the plurality of photos to the specific cluster and deleting the second specific photo from the specific cluster, adjusting, at the computing device, at least one of:
• (i) the quality metrics and (ii) the similarity matrix to obtain at least one of modified quality metrics and a modified similarity matrix, and
• (iii) the subset of the plurality of photos to obtain a first modified subset of the plurality of photos;
• selecting, at the computing device, a second modified subset of the plurality of photos by performing joint global maximization of photo quality and photo diversity using at least one of the modified quality metrics and the modified similarity matrix;
• determining, at the computing device, a second plurality of modified clusters for the second modified subset of the plurality of photos using the hierarchical clustering algorithm, each of the second plurality of modified clusters including one or more distinct photos from the second modified subset of the plurality of photos;
• and
• storing, at the computing device, the second modified subset of the plurality of photos arranged based on the second plurality of modified clusters.

A computing device comprising:

• an input/output device configured to receive a plurality of photos;
• one or more processors configured to:
• select a subset of the plurality of photos by performing joint global maximization of photo quality and photo diversity using a quality metric for each of the plurality of photos and a similarity matrix for the plurality of photos, the quality metric for a specific photo being indicative of a quality of the specific photo, the similarity matrix being indicative of a similarity between each unique pair of photos of the plurality of photos, and
• determine one or more clusters for the subset of the plurality of photos using a hierarchical clustering algorithm, the one or more clusters collectively including less than or equal to all of the subset of the plurality of photos, each of the one or more clusters including one or more distinct photos from the subset of the plurality of photos;
• and
• a memory configured to store the subset of the plurality of photos arranged based on the one or more clusters,
• wherein the input/output device is further configured to receive user input with respect to the subset of the plurality of photos arranged based on the one or more clusters, the user input corresponding to at least one of adding a first specific photo from a remainder of the plurality of photos to a specific cluster and deleting a second specific photo from the specific cluster,
• wherein when the user input corresponds to at least one of adding the first specific photo from the remainder of the plurality of photos to the specific cluster and deleting the second specific photo from the specific cluster, the one or more processors are further configured to adjust at least one of:
• (i) the quality metrics and (ii) the similarity matrix to obtain at least one of modified quality metrics and a modified similarity matrix, and
• (iii) the subset of the plurality of photos to obtain a first modified subset of the plurality of photos;
• select a second modified subset of the plurality of photos by performing joint global maximization of photo quality and photo diversity using at least one of the modified quality metrics and the modified similarity matrix;
• and
• determine a second plurality of modified clusters for the second modified subset of the plurality of photos using the hierarchical clustering algorithm, each of the second plurality of modified clusters including one or more distinct photos from the second modified subset of the plurality of photos, and
• the memory is further configured to store the second modified subset of the plurality of photos arranged based on the second plurality of modified clusters.