Marco Furini

STIMO: STIll and MOving video storyboard for the web scenario

In the current Web scenario a video browsing tool that produces on-the-fly storyboards is more and more a need. Video summary techniques can be helpful but, due to their long processing time, they are usually unsuitable for on-the-fly usage. Therefore, it is common to produce storyboards in advance, penalizing users customization. The lack of customization is more and more critical, as users have different demands and might access the Web with several different networking and device technologies. In this paper we propose STIMO, a summarization technique designed to produce on-the-fly video storyboards. STIMO produces still and moving storyboards and allows advanced users customization (e.g., users can select the storyboard length and the maximum time they are willing to wait to get the storyboard). STIMO is based on a fast clustering algorithm that selects the most representative video contents using HSV frame color distribution. Experimental results show that STIMO produces storyboards with good quality and in a time that makes on-the-fly usage possible.

Real-time traffic transmission over the Internet

Multimedia applications require the transmission of real-time streams over a network. These streams often exhibit variable bandwidth requirements, and require high bandwidths and guarantees from the network. This creates problems when such streams are delivered over the Internet. To solve these problems, recently, a small set of differentiated services has been introduced. Among these, Premium Service is suitable for transmitting real-time stored stream (full knowledge of the stream characteristics). It uses a bandwidth allocation mechanism (BAM) based on the stream peak rate. Due to the variable bandwidth requirement, the peak rate BAM can waste large amount of bandwidth. In this paper we propose a new BAM that uses less bandwidth than the peak rate BAM, while providing the same service. Our BAM does not affect the real-time stream quality of service (QoS) and does not require any modification to the Premium Service Architecture. We also introduce several frame dropping mechanisms that further reduce bandwidth consumption subject to a QoS constraint when coupled with the above BAM. The proposed BAM and the dropping mechanisms are evaluated using Motion JPEG and MPEG videos and are shown to be effective in reducing bandwidth requirements. Further, since VCR operations are very useful in video streaming, we propose a mechanism that introduces these operations in our BAM. Through simulations we show the effectiveness of this mechanism.

Location privacy and public metadata in social media platforms: attitudes, behaviors and opinions

The highavailability of geolocation technologies is changing the social media mobile scenario and is exposing users to privacy risks. Different studies have focused on location privacy in the mobile scenario, but the results are conflicting: some say that users are concerned about location privacy, others say they are not. In this paper, we initially investigate attitudes and behaviors of people toward a location-aware scenario; then, we show users the amount of personal and sensitive data that can be extracted from contents publicly available in social platforms, and finally we ask for their opinions about a location-aware scenario. Results show that people who were not initially concerned about privacy are the most worried about the location-aware scenario; conversely, people who were initially concerned are less worried about the location-aware scenario and find the scenario interesting. A deeper analysis of the obtained results allows us to draw guidelines that might be helpful to build an effective location-aware scenario.

VISTO: visual storyboard for web video browsing

Web video browsing is rapidly becoming a very popular activity in the Web scenario, causing the production of a concise video content representation a real need. Currently, static video summary techniques can be used to this aim. Unfortunately, they require long processing time and hence all the summaries are produced in advance without any users customization. With an increasing number of videos and with the large users heterogeneousness, this is a burden. In this paper we propose VISTO, a summarization technique that produces customized on-the-fly video storyboards. The mechanism uses a fast clustering algorithm that selects the most representative frames using their HSV color distribution and allows users to select the storyboard length and the processing time. An objective and subjective evaluation shows that the storyboards are produced with good quality and in a time that allows on-the-fly usage.

Users Behavior in Location-Aware Services: Digital Natives versus Digital Immigrants

Location-aware services may expose users to privacy risks as they usually attach user’s location to the generated contents. Different studies have focused on privacy in location-aware services, but the results are often conflicting. Our hypothesis is that users are not fully aware of the features of the location-aware scenario and this lack of knowledge affects the results. Hence, in this paper we present a different approach: the analysis is conducted on two different groups of users (digital natives and digital immigrants) and is divided into two steps: (i) understanding users’ knowledge of a location-aware scenario and (ii) investigating users’ opinion toward location-aware services after showing them an example of an effective location-aware service able to extract personal and sensitive information from contents publicly available in social media platforms. The analysis reveals that there is relation between users’ knowledge and users’ concerns toward privacy in location-aware services and also reveals that digital natives are more interested in the location-aware scenario than digital immigrants. The analysis also discloses that users’ concerns toward these services may be ameliorated if these services ask for users’ authorization and provide benefits to users. Other interesting findings allow us to draw guidelines that might be helpful in developing effective location-aware services.

An audio-video summarization scheme based on audio and video analysis

The availability of video files in the Internet is growing at an exceptional speed and in the near future video browsing will be a common activity. To facilitate such activity it will be necessary to have a small clip for any given video. Currently, video skimming and video summarization techniques can reduce the temporal representation of a given video. However, most of these techniques do not include audio in the produced summaries. Here, we propose a mechanism that, using audio and video analysis, produces video summaries coupled with intelligible audio. Experimental results show that the summaries are largely reduced (up to 50%) and that the perceived video quality may be comparable to the one of the original video (in term of jerkiness). Consumers satisfaction has been investigated through MOS and results show that our summaries can be considered as an alternative to the original videos.

Ad hoc networks: a protocol for supporting QoS applications

A delay-bounded service in wireless ad hoc networks is challenging, as ad hoc networks do not provide any type of guarantees. Several protocols have been proposed to support applications without timing requirements in ad hoc networks, but the increasing demand of QoS applications, in ad hoc wireless environments, requires delay-bound service. The contribution of this paper is to propose a protocol that provides QoS service, by means of timing guarantees, to the supported applications in ad hoc wireless networks.

TRank: Ranking Twitter users according to specific topics

Twitter is the most popular real-time micro-blogging service and it is a platform where users provide and obtain information at rapid pace. In this scenario, one of the biggest challenge is to find a way to automatically identify the most influential users of a given topic. Currently, there are several approaches that try to address this challenge using different Twitter signals (e.g., number of followers, lists, metadata), but results are not clear and sometimes conflicting. In this paper, we propose TRank, a novel method designed to address the problem of identifying the most influential Twitter users on specific topics identified with hashtags. The novelty of our approach is that it combines different Twitter signals (that represent both the user and the users tweets) to provide three different indicators that are intended to capture different aspects of being influent. The computation of these indicators is not based on the magnitude of the Twitter signals alone, but they are computed taking into consideration also human factors, as for example the fact that a user with many active followings might have a very noisy time lime and, thus, miss to read many tweets. The experimental assessment confirms that our approach provides results that are more reasonable than the one obtained by mechanisms based on the sole magnitude of data.

Bringing multimedia contents into MP3 files

The digital music revolution has improved the availability of music and made it easier to enjoy. The shift from hard support, such as audio CDs, to software leaves some unsatisfied by the loss of additional media contents, such as images, lyrics, and CD cover. We propose to fill the gap by enriching e-music with multimedia contents. Our proposal focuses specifically on MP3 as a widely distributed open format. The challenge we meet in enriching MP3 lies in maintaining audio compatibility and avoiding file size explosion. To this end, we define a description language to express multimedia contents in a textual way, with higher representation efficiency than existing languages. We also address security issues that arise from the use of an open format. Effectiveness is shown with an MP3 player capable of rendering multimedia contents.

Audio-text synchronization inside mp3 files: a new approach and its implementation

The large usage of multimedia portable devices has contributed to a rapid increase in the demand for multimedia entertainment services. We focus on the karaoke service: several systems have been proposed but they are too difficult to be used directly over audio devices. Conversely, we propose a very simple approach to providing a karaoke-like service over any audio device that can play out mp3 files. Our approach is essentially new, as no additional files, beyond the mp3 file, are necessary. A simple description language has been designed to characterize audio-text timing synchronizations, which are transparently stored inside the mp3 file. The effectiveness of our approach is proved through a developed Java mp3 player. The simplicity of our approach along with Java portability allow a straightforward use of our player over any OS and over any audio device that supports Java applications.