Automatic creation of photo books from stories in social media

Abstract

Photos are a special way to tell stories of our best memories and moments. The representation of those photos in appealing physical photo books is highly appreciated by many people. Today, many photos are shared via social networking sites, where people upload their photos and share their stories with their friends. The members of social networks comment on each other's photos, add tags or descriptions and upload new photos of the same events to their albums. While the media of different personal events are available on the social network, there is no easy way to collect and bundle them into a story and print this story as a photo book. We propose an approach to automatically detect media elements that match a query (where, when, what, who) in the user's social network and intelligently arrange and compose them into a printable photo book. We combine content analysis of text and images to automatically and semi-automatically select photos of a specific story. We calculate the probabilities of each two photos to belong to the same event using an Expectation-Maximization algorithm that we propose in order to be able to retrieve them easily when receiving the user queries, and we address the differences between our model and other models that use similar proposed algorithms. People's tags and the interaction between the users and the photos as well as other semantic information are exploited to select important photos that are suitable to create the photo book. The selected photos and derived semantics are then employed to automatically create an appealing layout for the photo book.

Large scale flexible event-based clustering from photos in social media

“…For example [2] introduced an unsupervised approach to cluster the photos to a known number of events using an expectation maximization algorithm, where the meta data, the content based and the camera based features were taken into account. More recently in [3] we also introduced an unsupervised algorithm that exploits social features as well as content features in the social networks in order to estimate the probabilistic relation between the photos. In the work of [4] the event was characterized in terms of four facets (Where, When, What, Who) and the user's social context was used to annotate photos because users do not always agree on annotations.…”

“…Friends may attend the same events and upload photos of each others in different accounts. For example, in [3] we showed that a large proportion of Facebook users have a high percentage of their photos added by others. Thus, it is not easy to find all photos belonging to one event by browsing a single album or account.…”

“…People often document such events in social media platforms by adding related media items such as videos or photos. By this they are able to keep a diary-like memory for themselves, to share it with friends or to use the selected media items later in a media presentation such as photo books [3] or video booklets [7]. However, it is generally not an easy task to automatically categorize the media items to their events, especially as the events can significantly vary temporally: E.g., a birthday party may take several hours, while a trip or a vacation may take few weeks.…”

“…The media items presenting an event and the major features which describe it are shaped interactively by the user in a light-weight online process. Only a rough estimation of the relation between the photos is calculated offline in an efficient way by extending our approach described in [3]. The rest of this paper is organized as follows: In Section 2 we give insight to some of the related work.…”

Memory preservation made prestigious but easy

“…These sources can also being integrated while offering interfaces to retrieve the photos from there or to offer special ways of generating photo books implementing both, text and photos into a well arranged product as shown in [10].…”