Hugo Feitosa de Figueirêdo

PhotoGeo: a photo digital library with spatial-temporal support and self-annotation

The recent popularity of digital cameras has posed a new problem: how to efficiently store and retrieve the very large number of digital photos captured and chaotically stored in multiple locations without any annotation. This paper proposes an infrastructure, called PhotoGeo, which aims at helping users with the people photo annotation, event photo annotation, storage and retrieval of personal digital photos. To achieve the desired objective, PhotoGeo uses new algorithms that make it possible to annotate photos with the key metadata to facilitate their retrieval, such as: the people who were shown in the photo (who); where it was captured (where); the date and time of capture (when); and the event that was captured. The paper concludes with a detailed evaluation of these algorithms.

Expandindo e utilizando informações de contexto para a sugestao de anotações de fotografias digitais

Digital photography is definitively becoming very popular due to several advantages over the analog one which include easy processing, copying, sharing and cheap storing. Moreover, storage devices for digital photos are getting cheaper and photo-capturing devices are no longer limited to digital cameras but include mobile phones, PDAs and others. As a consequence, there is a new challenge to efficiently annotate, store and retrieve these very large collections of digital photos captured. This paper addresses these issues and presents a new algorithm to help users to annotate geo-referenced photos, using user context and information from external sources. A digital library to store these photo collections and their metadata; and an experimental evaluation are also discussed.

VideoLib: a video digital library with support to spatial and temporal dimensions

Nowadays, a large amount of digital video can be found on the Web, and this number is increasing rapidly. Metadata should be associated to video in order to enable an efficient identification, indexing and semantic representation of it, so that effective video information retrieval tools can be built. Both Dublin Core and MPEG-7 standards provide a metadata set which enables the description of multimedia content. This paper presents a video digital library, known as VideoLib, which utilizes Dublin Core and MPEG-7 audiovisual descriptions to provide video content retrieval. Mainly, VideoLib enhances video retrieval by introducing spatial and temporal operators.

Detection of photos from the same event captured by distinct cameras

Currently one of the main features of events such as birthday party, wedding, conferences, and trips is the amount of photos taken from different digital cameras. The plethora of photo devices such as smartphones and tablets have contributed for this avalanche of photos. Social networks such as Facebook, Orkut, Flickr, and Picasa store a large volume of photos. As a consequence, the retrieval of such photos demands new systems. The automatic annotation of photo context metadata such as timestamp, and geographic coordinates, represents a challenge. In this paper, we propose a method for automatic event detection based on photos taken from different cameras in a same event. The proposed method aims to improve the retrieval of photos in a same event, to make cross tagging and to detect errors in the annotations.

SocialSupervisor: a geographically enhanced social content site to supervise public works

In the Web 2.0 users may freely contribute with information on several topics. Hence, public agencies can take advantage of this fact by creating communication channels with society. This paper presents the SocialSupervisor, a geographically enhanced social content site, where citizens can be notified about public works nearby. Moreover, they are able to post complaints related to any irregularity in public services. Such information forces the government to promote an immediate investigation of the facts.

A context-aware solution to annotate people in mobile devices

Recently, digital photography is definitely becoming very popular. Thus, there is a large increase in the number of captured photos, which demands better methods to manage this data. An approach is to allow the user to annotate some information in those photos, for instance, the people who appear in them. In this paper, two solutions are proposed in order to assist the user in the process of annotating people in photos, using mobile devices. The first one, called on-line, accesses a server to produce the results and the other one, called off-line, utilizes the mobile device itself to compute the results. Finally, a comparative study between the solutions, using parameters such as hit rate and performance, is presented.

Photo annotation: a survey

Due to the large number of photos that are currently being generated, it is very important to have techniques to organize, search for, and retrieve such images. Photo annotation plays a key role in these mechanisms because it can link raw data (photos) to specific information that is essential for human beings to handle large amounts of content. However, the generation of photo annotation is still a difficult problem to solve as part of a well-known challenge called the semantic gap. In this paper, a literature review was conducted with the aim of investigating the most popular methods employed to produce photo annotations. Based on the papers surveyed, we identified that People (“Who?”), Location (“Where?”), and Event (“Where? When?”) are the most important features of photo annotation. We also established comparisons between similar photo annotation methods, highlighting key aspects of the most commonly used approaches. Moreover, we provide an overview of a general photo annotation process and present the main aspects of photo annotation representation comprising formats, context of usage, advantages and disadvantages. Finally, we discuss ways to improve photo annotation methods and present some future research guidelines.

Semi-automatic photograph tagging by combining context with content-based information

We propose a semi-automatic technique for the annotation of people in photographs.The technique uses context and content information and is based on the weighted sum of estimators.Multivariable linear regression and slope functions are adopted to weight the estimators.The technique obtained better results than the context-based and content-based methods when the metric H-Hit rate was used. This article proposes a semi-automatic technique for the annotation of people in photographs. The technique uses context and content information and is based on a weighted sum of estimators, which results in a list of the persons contacts that are more likely to be present in a photograph. Machine learning methods, such as multivariable linear regression and slope function, are adopted to filter and weight the estimators and eigenfaces for face recognition. The article also describes the results of experiments that were performed with a collection of 4050 photographs with 365 different people, which indicate that the proposed technique outperforms techniques that adopt only context or only content using as a performance metric the H-Hit rate of correct annotations.

New Approaches for Geographic Location Propagation in Digital Photograph Collections

The integration of GPS in smartphones, tablets and digital cameras becomes more present, resulting in a large amount of multimedia files. As GPS receivers may not work well indoors, this problem may generate incorrect locations, very distant from the real location where the picture was taken, or even generate no location at all. So, to deal with these inconsistencies, this work proposes two novel location propagation techniques. These techniques were validated through a comparative analysis with two other techniques. Some metrics were used to validate the techniques: precision, recall and accuracy in the photographs location propagation. The results prove that the correct choice for location propagation technique depends on the importance of each metric and on the system user profile. Besides the choice of the correct technique, we also show that the order of the photographs that will receive the location propagation must be random.

Towards Better Propagation of Geographic Location in Digital Photo Collections

The integration of GPS in smartphones, tablets and digital cameras become increasingly popular, but GPS receivers does not work well indoors. This malfunction can generate erroneous location information from where the picture was really taken, or no information at all. In order to overcome this problem, this paper proposes an automatic selection of techniques that couples different solutions to the problem from multiple regression. The focus of the work is to minimize the error generated by existing techniques with an automatic selection of techniques that uses Artificial Intelligence methods to automatically select the best technique. Also, we execute an experiment to validate the results achieved. After validation, the tests indicated that the proposed automatic selection technique improved the results in five of the six scenarios. In the sixth scenario the results were the same.